Ewing Sarcoma: Methods and Protocols 1071610198, 9781071610190

This volume aims at providing a selection of the most effective and widely employed protocols used in the Ewing sarcoma

238 63 10MB

English Pages 337 [333] Year 2021

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Preface
Contents
Contributors
Part I: Molecular Biology Techniques Applied To Ewing Sarcoma
Chapter 1: Germline Variation and Somatic Alterations in Ewing Sarcoma
1 Introduction
2 Identifying a DNA Source for Investigation
3 Considerations When Selecting EwS Cases and Controls
4 Selection of Optimal Genomic Technologies
5 Key Considerations in EwS Germline Association Analyses
6 Analysis Elements for NGS Projects
7 Final Considerations for EwS Genetic Investigations
References
Chapter 2: Western Blot Analysis in Ewing Sarcoma
1 Introduction
1.1 Western Blot, the Technique
1.2 Ewing Sarcoma Specifications
2 Materials
2.1 Reagents
2.2 Equipment
3 Methods
3.1 Sample Preparation
3.2 SDS-PAGE Gel Electrophoresis
3.3 Protein Transference
3.4 Antibody Staining
3.5 Analysis
4 Notes
References
Chapter 3: Tissue Preservation and FFPE Samples: Optimized Nucleic Acids Isolation in Ewing Sarcoma
1 Introduction
2 Materials
2.1 Fresh Tissue Sample Preservation
2.2 Frozen Tissue Sample Preservation
2.3 FFPE Tissue Sample Preservation
2.4 Optimized DNA/RNA Isolation from FFPE Samples
3 Methods
3.1 Fresh Tissue Sample Preservation
3.2 Frozen Tissue Specimen Preservation
3.2.1 Long-Term Cold Storage
3.2.2 Snap Freezing in Liquid Nitrogen
3.3 FFPE Sample Preservation
3.3.1 FFPE Sample Preparation
3.3.2 FFPE Samples Issues for Molecular Analysis
3.4 Optimized DNA/RNA Isolation from FFPE Samples
3.4.1 Tumor Cells Enrichment
3.4.2 Deparaffination
3.4.3 DNA and RNA Isolation
3.4.4 Measurement of the Concentration and Quality of Nucleic Acids
4 Notes
References
Chapter 4: Liquid Biopsies in Ewing Sarcoma
1 Introduction
2 Materials
2.1 Plasma Sample Collection and Storage
2.2 Isolation of Cell-Free Circulating DNA
2.3 Quantification of ctDNA in Ewing sarcoma Patients
3 Methods
3.1 Plasma Sample Collection and Storage
3.2 Isolation of Cell-Free Circulating DNA
3.3 Quantification of ctDNA in Ewing Sarcoma Patients
4 Notes
References
Part II: Morpho-Molecular Diagnostics in Ewing Sarcoma
Chapter 5: (Immuno)histological Analysis of Ewing Sarcoma
1 Introduction
2 Materials
3 Methods
3.1 Grossing
3.2 Fixation
3.3 Decalcification
3.4 Pathological Diagnosis and Assessment of Histological Response to Neoadjuvant Therapy
3.5 Immunohistochemistry and Interpretation
3.6 Reporting Template of a Core Biopsy with EwS
3.7 Reporting Template of a Surgical Resection Ewing Sarcoma
4 Notes
References
Chapter 6: Molecular Approaches to Diagnosis in Ewing Sarcoma: Fluorescence In Situ Hybridization (FISH)
1 Introduction
2 Materials
3 Methods
3.1 Technical Procedure
3.2 Microscopic Evaluation and Interpretation
4 Notes
References
Chapter 7: Molecular Approaches to Diagnosis in Ewing Sarcoma: RT-PCR
1 Introduction
2 Materials
2.1 RNA Extraction
2.2 Reverse Transcription (RT)
2.3 Polymerase Chain Reaction (PCR)
2.4 Agarose Gel Electrophoresis
3 Methods
3.1 RNA Extraction
3.2 Reverse Transcription
3.3 Polymerase Chain Reation (PCR)
3.4 Gel Electrophoresis
3.5 Interpretation of the Results
4 Notes
References
Chapter 8: Molecular Approaches to Diagnosis in Ewing Sarcoma: Targeted RNA Sequencing
1 Introduction
2 Materials
2.1 Total Nucleic Acid Extraction
2.2 RNA Quantification
2.3 Library Preparation, Quantification, Normalization, and Sequencing
2.4 Data Analysis
3 Methods
3.1 Extraction of Total Nucleic Acid from FFPE Samples
3.2 RNA Quantification
3.3 Library Preparation
3.4 Library Quantification, Normalization, and Pooling
3.5 Library Sequencing
3.6 Data Analysis
4 Notes
References
Part III: Cellular Biology and Cell Culture Techniques
Chapter 11: Proliferation Assessment by Trypan Blue Exclusion in Ewing Sarcoma
1 Introduction
2 Materials
3 Methods
3.1 Preparation of Cells and Seeding
3.2 Experimental Procedure
3.3 Readout
4 Notes
References
Chapter 12: Drug Screening by Resazurin Colorimetry in Ewing Sarcoma
1 Introduction
2 Materials
3 Methods
3.1 Preparation and Seeding of Cells
3.2 Stimulation with Compounds
3.3 Resazurin Assay
3.4 Analysis
4 Notes
References
Chapter 13: Analysis of Migration and Invasion in Ewing Sarcoma
1 Introduction
2 Materials
2.1 Wound Healing Assay
2.2 Transwell Assay
3 Methods
3.1 Wound Healing Assay
3.1.1 Experimental Procedure
3.1.2 Wound Healing Analysis
3.2 Transwell Assay
3.2.1 Migration Analysis
3.2.2 Invasion Analysis
3.2.3 Transwell Analysis
4 Notes
References
Chapter 9: Ewing Sarcoma-Specific (Re)expression Models
1 Introduction
2 Materials
2.1 Vector Design
2.2 Cloning of the Transfer Plasmid
2.3 Lentivirus Production
2.4 Titering of Viral Particles Using Flow Cytometry
2.5 Titering of Viral Particles Using qPCR
2.6 Transduction and Selection
2.7 Validation of Gene Expression
3 Methods
3.1 Vector Design
3.2 Cloning of Transfer Plasmids
3.3 Lentivirus Production
3.4 Titering of Viral Particles Using Flow Cytometry
3.5 Titering of Viral Particles Using qPCR
3.6 Viral Transduction and Selection
3.7 Validation of Gene Expression
4 Notes
References
Chapter 10: Analysis of Regulatory DNA Sequences by Dual-Luciferase Reporter Assays in Ewing Sarcoma
1 Introduction
2 Materials
3 Methods
3.1 Cloning of Reporter Plasmids
3.2 Transfection of Ewing Sarcoma Cells
3.3 Assessing Luciferase Expression
3.4 Data Analysis
4 Notes
References
Part IV: Animal Models in Ewing Sarcoma
Chapter 14: Genetically Engineered Mouse Model in Ewing Sarcoma
1 Introduction
2 Materials
2.1 Animals
2.2 Cell Culture
2.3 Plasmids
2.4 Antibodies
3 Methods
3.1 Preparation of Mouse Embryonic Superficial Zone (eSZ) Cells
3.2 Retroviral Infection
3.3 Transplantation
4 Notes
References
Chapter 15: Tumor Growth Analysis of Ewing Sarcoma Cell Lines Using Subcutaneous Xenografts in Mice
1 Introduction
2 Materials
3 Methods
3.1 Preparation of Tumor Cell Suspension
3.2 Subcutaneous Injection of Tumor Cells
3.3 Monitoring of Tumor Growth
3.4 Extraction of Xenografted Tumor Tissue
4 Notes
References
Chapter 16: Metastasis Assessment in Ewing Sarcoma Using Orthotopic Xenografts
1 Introduction
2 Materials
3 Methods
3.1 Cell Preparation
3.2 Cells Inoculation in the Hind Limb
3.3 Tumor Development
3.3.1 Monitor of Tumor Growth by Caliper Measurement
3.3.2 IVIS Lecture Monitoring of Tumor Growth
3.3.3 Drug Administration
3.4 Surgery
3.5 Follow Up of Mice at Long Term
3.6 End Point: Euthanization of Mice
3.7 Histological Analysis
3.8 Metastasis Detection
3.9 Final Analysis of the Data
4 Notes
References
Chapter 17: Orthotopic Implants in Mice
1 Introduction
2 Materials
2.1 Tumor Dissociation
2.2 Preparation of Cell Suspension
2.3 Orthotopic Injection
3 Methods
3.1 Tumor Dissociation
3.2 Preparation of Cell Suspension
3.3 Orthotopic Injection
4 Notes
References
Chapter 18: Ewing Sarcoma PDX Models
1 Introduction
2 Materials
2.1 Tissue Collection and Preparation
2.2 Tumor Engraftment
2.3 Passages and Tumor Collection from Mice
2.4 Ewing Sarcoma PDX Validation
2.4.1 Histology (See Note 2)
2.4.2 Immunohistochemistry (See Note 2)
2.4.3 PDX Chimerism
3 Methods
3.1 Tissue Collection, Preparation and Delivery of Fresh and Frozen Samples
3.2 Tumor Engraftment Over or Under the Inter-Scapular Brown Fat Pad
3.3 Passages and Tumor Collection from Mice
3.4 Histological Validation
3.5 Immunohistochemical Validation
3.5.1 Deparaffinization/Rehydration
3.5.2 Antigen Unmasking (if Necessary)
3.5.3 Inhibition of Endogenous Peroxidase
3.5.4 Immunostaining
3.5.5 Development of Immunoreaction
3.5.6 Dehydration
3.6 Ewing Sarcoma PDX: Evaluating Human and Murine Chimerism
3.6.1 Isolation of RNA
3.6.2 Reverse Transcription
3.6.3 Quantitative PCR
3.7 Ewing Sarcoma PDX: STR Profiling
3.8 Ewing Sarcoma Comprehensive Molecular Characterization
4 Notes
References
Chapter 19: Using Zebrafish Larvae as a Xenotransplantation Model to Study Ewing Sarcoma
1 Introduction
2 Materials
2.1 Cell Preparation
2.2 Fish Breeding and Maintenance
2.3 Xenografting
2.4 Imaging
2.4.1 Manual Imaging
2.4.2 Automated Imaging/Quantification
2.5 Immunostaining
3 Methods
3.1 Overview
3.2 Preparing shSK-E17T Ewing Sarcoma Cells for Xenotransplantation
3.2.1 Preparation of shSK-E17T Cell Solution
3.2.2 Alternative Labeling Strategy with DiI
3.3 Preparing Zebrafish Embryos for Xenografting
3.4 Xenografting shSK-E17T Ewing sarcoma Cells into 2 dpf Zebrafish Embryos
3.5 Analysis of Ewing Sarcoma Xenografts
3.5.1 Manual Imaging of Xenotransplanted Zebrafish Larvae
3.5.2 Automated Image Acquisition of Xenotransplanted Zebrafish Larvae Using a High Content Imager
3.5.3 Immunostaining and Confocal Analysis of Xenotransplanted Zebrafish Larvae
4 Notes
References
Part V: Bioinformatic Approaches to Ewing Sarcoma
Chapter 20: Main Repositories and Databases Used in Ewing Sarcoma Research
1 The Gene Expression Omnibus (GEO)
2 ArrayExpress
3 cBioPortal
4 OnkoKB
5 The Cancer Genome Atlas (TCGA) Research Network
6 AACR GENIE
7 FusionHub
References
Chapter 21: Chromatin Immunoprecipitation Followed by Next-Generation Sequencing (ChIP-Seq) Analysis in Ewing Sarcoma
1 Introduction
1.1 ChIP-Seq: A Powerful Tool to Study Protein-DNA Interactions
1.2 Applications of ChIP-seq in Cancer Research
1.3 ChIP-seq Contributed to Important Breakthrough in Ewing Sarcoma
2 Materials
2.1 Computing Requirements
2.2 Software Installation
2.3 Data
3 Methods
3.1 Sample Collection and Processing
3.2 Quality Control (QC) and Read Filtering
3.3 Mapping Using BWA
3.4 Peak Calling Using HMcan
3.5 Between-Sample Normalization of ChIP-seq Data
3.6 Subsequent Analysis
3.6.1 Motif Analysis
3.6.2 Annotation of Binding Sites
3.6.3 Super Enhancer (SE) Calling
3.6.4 Conversion Between Genome Builds
4 Notes
References
Chapter 22: Epigenetic Analysis in Ewing Sarcoma
1 Introduction
2 Materials/Applications
3 Methods
3.1 Data Acquisition and Preprocessing
3.2 Alignment of Data
3.3 Peak Calling
3.4 Visualization
3.5 CREATE HEATMAPS
3.6 Transcription Factor Motif Enrichment
3.7 Genomic Feature Association of Peaks
4 Notes
References
Chapter 23: Systems Biology Analysis for Ewing Sarcoma
1 Introduction
2 Omics and Large-Scale Perturbation Studies of EwS
2.1 Sources of Public Omics Datasets
2.2 Genomic Studies of EwS
2.2.1 Microarray-Based Comparative Genomic Hybridization Studies
2.2.2 Genomewide Association Studies (GWAS) of EwS
2.2.3 Sequencing EwS Genome, WES and WGS Approaches
2.3 Studies of EwS Transcriptome
2.3.1 Expression of Coding Genes in EwS
2.3.2 Expression of Noncoding Parts of the Genome: microRNA and lncRNA
2.4 Characterizing the EwS´s Epigenome
2.5 Proteomic Studies
2.6 Metabolomic Studies
2.7 Systematic Perturbation Studies, Drug and Gene Invalidation Screenings
3 Computational Systems Biology Studies of EwS
3.1 Multi-omics EwS Datasets
3.2 Integrative Biology Studies Combining Several Data Types and Sources
3.3 Assessing EwS Tumor Composition Via Mathematical Modeling
3.4 Network Modeling Approaches
4 Single Cell Studies of EwS
5 Conclusion
References
Index
Recommend Papers

Ewing Sarcoma: Methods and Protocols
 1071610198, 9781071610190

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Methods in Molecular Biology 2226

Florencia Cidre-Aranaz Thomas G. P. Grünewald Editors

Ewing Sarcoma Methods and Protocols

METHODS

IN

MOLECULAR BIOLOGY

Series Editor John M. Walker School of Life and Medical Sciences University of Hertfordshire Hatfield, Hertfordshire, UK

For further volumes: http://www.springer.com/series/7651

For over 35 years, biological scientists have come to rely on the research protocols and methodologies in the critically acclaimed Methods in Molecular Biology series. The series was the first to introduce the step-by-step protocols approach that has become the standard in all biomedical protocol publishing. Each protocol is provided in readily-reproducible step-bystep fashion, opening with an introductory overview, a list of the materials and reagents needed to complete the experiment, and followed by a detailed procedure that is supported with a helpful notes section offering tips and tricks of the trade as well as troubleshooting advice. These hallmark features were introduced by series editor Dr. John Walker and constitute the key ingredient in each and every volume of the Methods in Molecular Biology series. Tested and trusted, comprehensive and reliable, all protocols from the series are indexed in PubMed.

Ewing Sarcoma Methods and Protocols

Edited by

Florencia Cidre-Aranaz and Thomas G. P. Grünewald Division of Translational Pediatric Sarcoma Research, German Cancer Research Center (DKFZ) Hopp-Children’s Cancer Center (KiTZ), Heidelberg, Germany

Editors Florencia Cidre-Aranaz Division of Translational Pediatric Sarcoma Research German Cancer Research Center (DKFZ) Hopp-Children’s Cancer Center (KiTZ) Heidelberg, Germany

Thomas G. P. Gru¨newald Division of Translational Pediatric Sarcoma Research German Cancer Research Center (DKFZ) Hopp-Children’s Cancer Center (KiTZ) Heidelberg, Germany

ISSN 1064-3745 ISSN 1940-6029 (electronic) Methods in Molecular Biology ISBN 978-1-0716-1019-0 ISBN 978-1-0716-1020-6 (eBook) https://doi.org/10.1007/978-1-0716-1020-6 © Springer Science+Business Media, LLC, part of Springer Nature 2021 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Humana imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer Nature. The registered company address is: 1 New York Plaza, New York, NY 10004, U.S.A.

Preface Ewing sarcoma is a highly aggressive bone and soft tissue tumor, which mainly affects children, adolescents, and young adults. It is a clinically complex disease, as despite treatment by local resection and systemic chemotherapy, patients presenting with metastasis at diagnosis or relapse have still a dismal prognosis. Paradoxically, Ewing sarcoma is a genetically relatively simple cancer, characterized by a pathognomonic fusion gene (EWSR1-ETS) that, once established, completely rewires the epigenetic and transcriptional signatures of the tumors’ cell(s)-of-origin. Although great advances have been achieved in the past decades, the exact mechanisms underlying Ewing sarcoma tumorigenesis and progression are still largely unknown. One of the most challenging aspects of Ewing sarcoma research and clinical assessment is the scarcity of well-curated, highly standardized, and optimally working methods. Thus, this volume of Methods in Molecular Biology™ aims at comprising a selection of the most effective and widely employed protocols in the Ewing sarcoma field. In order to accommodate a wide variety of in vitro, in vivo, and in silico techniques that may be applied to the analysis of Ewing sarcoma, this volume of the Methods in Molecular Biology™ series is organized into five sections: Section I deals with a selection of molecular biology techniques that have proven to be very helpful when analyzing Ewing sarcoma. This section starts with an introductory chapter explaining the key question of somatic alteration and genetic predisposition, which may aid to stratify patients according to risk in Ewing sarcoma. Additionally, this section describes Ewing sarcoma-specific protocols for molecular techniques including tissue preservation, nucleic acid isolation from formalin-fixed paraffinembedded samples, which are usually the result of patient’s solid biopsies and Western blotting. This chapter is complemented with a final comprehensive chapter on liquid biopsies. In order to help Ewing sarcoma clinicians and physician-scientists navigate the intricacies of diagnosis, Section II comprises all techniques applied to the morpho-molecular diagnostics in Ewing sarcoma, including a chapter on routine immune-histological analysis which additionally describes all pre- and post-diagnostic procedures and three subsequent chapters on every molecular aspect that can be part of Ewing sarcoma diagnosis or confirmation (fluorescence in situ hybridization (FISH), RT-PCR, and targeted RNA sequencing). Section III of this volume describes a collection of in vitro tools and cell culture functional assays to guide researchers on essential wet-lab techniques from Ewing sarcoma-specific re-expression models, to luciferase assays adapted to the analysis of regulatory sequences present in Ewing sarcoma, to a set of functional assays implemented to evaluate the potential effect of re-expression or downregulation of particular genes or pathways and/or determine the effectiveness of selected drugs. Despite constant efforts since its first description in 1921, Ewing sarcoma remains a tumor entity of unclear cell of origin, which has underlined the importance of developing diverse methods that could recapitulate the disease in vivo. Section IV presents six chapters addressing these important animal models, starting with a method for a genetically engineered mouse model and covering other models such as classical subcutaneous xenografting and orthotopic implants. Due to the increasing need for systems that could better

v

vi

Preface

recapitulate the complexity of patient´s tumors, we have included in this Section IV a dedicated chapter on patient-derived xenografts (PDX) and orthotopic xenografts for metastasis assessment in Ewing sarcoma. In addition to these mouse models, a protocol for developing a zebrafish model is also detailed. As a result of constant development of new projects, Ewing sarcoma researchers create a staggering increase of highly valuable data. However, these data usually require being properly bioinformatically analyzed in order to obtain its maximum profitability. Thus, Section V of this volume contains four bioinformatic-focused chapters that range from a thorough compilation and description of repositories and databases essential for Ewing sarcoma research to a detailed explanation on how to address ChIP-seq, epigenetic, and systems biology analyses by using both own or already published data. Collectively, this volume of Methods in Molecular Biology™ will be of interest to molecular biologists and physician-scientists working in the field of Ewing sarcoma or in pediatric fusion-driven tumors, physicians interested in diagnostic aspects, as well as to bioinformaticians approaching the analysis of Ewing sarcoma. It has been our great pleasure to have collaborated in this volume with so many talented authors that are leaders in their respective fields. We are immensely thankful for their dedication to producing comprehensive chapters filled with extensively detailed tricks perfected at their laboratories, which will definitely help guide both the advanced and novice researchers. It is our hope that this volume comprising a collective technical knowledge around Ewing sarcoma will bring us one step forward in the quest to fight this devastating disease. Heidelberg, Germany

Florencia Cidre-Aranaz ¨ newald Thomas G. P. Gru

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

PART I

MOLECULAR BIOLOGY TECHNIQUES APPLIED TO EWING SARCOMA

1 Germline Variation and Somatic Alterations in Ewing Sarcoma . . . . . . . . . . . . . . . ¨ newald Mitchell J. Machiela and Thomas G. P. Gru 2 Western Blot Analysis in Ewing Sarcoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Aruna Marchetto and Laura Romero-Pe´rez 3 Tissue Preservation and FFPE Samples: Optimized Nucleic Acids Isolation in Ewing Sarcoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ¨ newald Laura Romero-Pe´rez and Thomas G. P. Gru 4 Liquid Biopsies in Ewing Sarcoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Manuela Krumbholz and Markus Metzler

PART II

v ix

3 15

27 39

MORPHO-MOLECULAR DIAGNOSTICS IN EWING SARCOMA

5 (Immuno)histological Analysis of Ewing Sarcoma. . . . . . . . . . . . . . . . . . . . . . . . . . . 49 ¨ newald, Antonio David Marcilla, Isidro Machado, Thomas G. P. Gru ´ lava Llombart-Bosch, and Enrique de A 6 Molecular Approaches to Diagnosis in Ewing Sarcoma: Fluorescence In Situ Hybridization (FISH) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Marcel Trautmann and Wolfgang Hartmann 7 Molecular Approaches to Diagnosis in Ewing Sarcoma: RT-PCR . . . . . . . . . . . . . 85 Carlos Rodrı´guez-Martı´n and Javier Alonso 8 Molecular Approaches to Diagnosis in Ewing Sarcoma: Targeted RNA Sequencing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Carmen Salguero-Aranda and Juan Diaz-Martin

PART III

CELLULAR BIOLOGY AND CELL CULTURE TECHNIQUES

9 Ewing Sarcoma-Specific (Re)expression Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . Maximilian M. L. Knott and Florencia Cidre-Aranaz 10 Analysis of Regulatory DNA Sequences by Dual-Luciferase Reporter Assays in Ewing Sarcoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tilman L. B. Ho¨lting and Maximilian M. L. Knott 11 Proliferation Assessment by Trypan Blue Exclusion in Ewing Sarcoma . . . . . . . . Cornelius Maximilian Funk and Julian Musa 12 Drug Screening by Resazurin Colorimetry in Ewing Sarcoma . . . . . . . . . . . . . . . . Julian Musa and Florencia Cidre-Aranaz

vii

119

139 151 159

viii

13

Contents

Analysis of Migration and Invasion in Ewing Sarcoma . . . . . . . . . . . . . . . . . . . . . . . 167 Florencia Cidre-Aranaz

PART IV ANIMAL MODELS IN EWING SARCOMA 14 15

16 17 18

19

Genetically Engineered Mouse Model in Ewing Sarcoma . . . . . . . . . . . . . . . . . . . . Miwa Tanaka and Takuro Nakamura Tumor Growth Analysis of Ewing Sarcoma Cell Lines Using Subcutaneous Xenografts in Mice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Florencia Cidre-Aranaz and Shunya Ohmura Metastasis Assessment in Ewing Sarcoma Using Orthotopic Xenografts . . . . . . . Roser Lo pez-Alemany and Oscar M. Tirado Orthotopic Implants in Mice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Elizabeth Stewart Ewing Sarcoma PDX Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Didier Surdez, Lorena Landuzzi, Katia Scotlandi, and Maria Cristina Manara Using Zebrafish Larvae as a Xenotransplantation Model to Study Ewing Sarcoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Susana Pascoal, Sarah Grissenberger, Eva Scheuringer, Rita Fior, Miguel Godinho Ferreira, and Martin Distel

PART V 20 21

22 23

183

191 201 215 223

243

BIOINFORMATIC APPROACHES TO EWING SARCOMA

Main Repositories and Databases Used in Ewing Sarcoma Research. . . . . . . . . . . Maria Debiec-Rychter Chromatin Immunoprecipitation Followed by Next-Generation Sequencing (ChIP-Seq) Analysis in Ewing Sarcoma . . . . . . . . . . . . . . . . . . . . . . . . . Gwenneg Kerdivel and Valentina Boeva Epigenetic Analysis in Ewing Sarcoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jeremy M. Simon and Nicholas C. Gomez Systems Biology Analysis for Ewing Sarcoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Marianyela Petrizzelli, Jane Merlevede, and Andrei Zinovyev

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

259

265 285 303 335

Contributors JAVIER ALONSO • Unidad de Tumores Solidos Infantiles, Instituto de Investigacion de Enfermedades Raras, Instituto de Salud Carlos III, Madrid, Spain; Centro de Investigacion Biome´dica en Red de Enfermedades Raras, Instituto de Salud Carlos III (CB06/07/1009; CIBERER-ISCIII), Madrid, Spain VALENTINA BOEVA • INSERM, U1016, Cochin Institute, CNRS UMR8104, Paris Descartes University, Paris, France; Department of Computer Science, ETH Zurich, Institute for Machine Learning, Zurich, Switzerland; Swiss Institute of Bioinformatics (SIB), Zu¨rich, Switzerland FLORENCIA CIDRE-ARANAZ • Division of Translational Pediatric Sarcoma Research, German Cancer Research Center (DKFZ), Heidelberg, Germany; Hopp Children’s Cancer Center (KiTZ), Heidelberg, Germany; German Cancer Consortium (DKTK), Partner Site Heidelberg, Heidelberg, Germany ´ LAVA • Institute of Biomedicine of Sevilla (IBiS), Virgen del Rocio University ENRIQUE DE A Hospital/CSIC/University of Sevilla/CIBERONC, Seville, Spain; Department of Normal and Pathological Cytology and Histology, School of Medicine, University of Seville, Seville, Spain MARIA DEBIEC-RYCHTER • Department of Human Genetics, KU Leuven an, University Hospital Leuven, Leuven, Belgium JUAN DIAZ-MARTIN • Department of Pathology, Instituto de Biomedicina de Sevilla, CSICUniversidad de Sevilla, Hospital Universitario Virgen del Rocı´o, Seville, Spain MARTIN DISTEL • Innovative Cancer Models, St. Anna Children’s Cancer Research Institute, Vienna, Austria MIGUEL GODINHO FERREIRA • Institute for Research on Cancer and Aging of Nice (IRCAN), Universite´ Coˆte d’Azur, Nice Cedex 2, France RITA FIOR • Champalimaud Centre for the Unknown, Champalimaud Foundation, Lisbon, Portugal CORNELIUS MAXIMILIAN FUNK • Division of Translational Pediatric Sarcoma Research, German Cancer Research Center (DKFZ), Heidelberg, Germany; Hopp Children’s Cancer Center (KiTZ), Heidelberg, Germany NICHOLAS C. GOMEZ • Laboratory of Mammalian Cell Biology and Development, The Rockefeller University, New York, NY, USA SARAH GRISSENBERGER • Innovative Cancer Models, St. Anna Children’s Cancer Research Institute, Vienna, Austria THOMAS G. P. GRU¨NEWALD • Division of Translational Pediatric Sarcoma Research, German Cancer Research Center (DKFZ), Hopp Children’s Cancer Center (KiTZ), Heidelberg, Germany WOLFGANG HARTMANN • Division of Translational Pathology, Gerhard-Domagk-Institute of Pathology, Mu¨nster, Germany TILMAN L. B. HO¨LTING • Max-Eder Research Group for Pediatric Sarcoma Biology, Faculty of Medicine, Institute of Pathology, LMU Munich, Munich, Germany GWENNEG KERDIVEL • Cochin Institute, INSERM U1016, CNRS UMR8104, University of Paris, Paris, France

ix

x

Contributors

MAXIMILIAN M. L. KNOTT • Max-Eder Research Group for Pediatric Sarcoma Biology, Faculty of Medicine, Institute of Pathology, LMU Munich, Munich, Germany MANUELA KRUMBHOLZ • Department of Pediatrics, University Hospital Erlangen, Erlangen, Germany LORENA LANDUZZI • IRCCS—Istituto Ortopedico Rizzoli, Experimental Oncology Laboratory, Bologna, Italy ANTONIO LLOMBART-BOSCH • Department of Pathology, University of Valencia, Valencia, Spain ROSER LO´PEZ-ALEMANY • Sarcoma Research Group, Oncobell Program, Bellvitge Biomedical Research Institute (IDIBELL), L’Hospitalet de Llobregat, Barcelona, Spain ISIDRO MACHADO • Pathology Department, Instituto Valenciano de Oncologı´a, Valencia, Spain; Pathology Department, Patologika Hospital Quiro n Valencia, Valencia, Spain MITCHELL J. MACHIELA • Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD, USA MARIA CRISTINA MANARA • IRCCS—Istituto Ortopedico Rizzoli, Experimental Oncology Laboratory, Bologna, Italy ARUNA MARCHETTO • Max-Eder Research Group for Pediatric Sarcoma Biology, Faculty of Medicine, Institute of Pathology, LMU Munich, Munich, Germany DAVID MARCILLA • Institute of Biomedicine of Sevilla (IBiS), Virgen del Rocio University Hospital/CSIC/University of Sevilla/CIBERONC, Seville, Spain JANE MERLEVEDE • Institut Curie, PSL Research University, Paris, France; INSERM U900, Paris, France; CBIO-Centre for Computational Biology, Mines ParisTech, PSL Research University, Paris, France MARKUS METZLER • Department of Pediatrics, University Hospital Erlangen, Erlangen, Germany JULIAN MUSA • Division of Translational Pediatric Sarcoma Research, German Cancer Research Center (DKFZ), Heidelberg, Germany; Department of General, Visceral and Transplantation Surgery, University of Heidelberg, Heidelberg, Germany; Hopp Children’s Cancer Center (KiTZ), Heidelberg, Germany; German Cancer Consortium (DKTK), Partner Site Heidelberg, Heidelberg, Germany TAKURO NAKAMURA • Division of Carcinogenesis, The Cancer Institute, Japanese Foundation for Cancer Research, Tokyo, Japan SHUNYA OHMURA • Max-Eder Research Group for Pediatric Sarcoma Biology, Institute of Pathology, Faculty of Medicine, LMU Munich, Munich, Germany SUSANA PASCOAL • Innovative Cancer Models, St. Anna Children’s Cancer Research Institute, Vienna, Austria MARIANYELA PETRIZZELLI • Institut Curie, PSL Research University, Paris, France; INSERM U900, Paris, France; CBIO-Centre for Computational Biology, Mines ParisTech, PSL Research University, Paris, France CARLOS RODRI´GUEZ-MARTI´N • Unidad de Tumores Solidos Infantiles, Instituto de Investigacion de Enfermedades Raras, Instituto de Salud Carlos III, Madrid, Spain LAURA ROMERO-PE´REZ • Division of Translational Pediatric Sarcoma Research, German Cancer Research Center (DKFZ), Heidelberg, Germany; Hopp Children’s Cancer Center (KiTZ), Heidelberg, Germany; German Cancer Consortium (DKTK), Heidelberg, Germany CARMEN SALGUERO-ARANDA • Department of Pathology, Instituto de Biomedicina de Sevilla, CSIC-Universidad de Sevilla, Hospital Universitario Virgen del Rocı´o, Seville, Spain

Contributors

xi

EVA SCHEURINGER • Innovative Cancer Models, St. Anna Children’s Cancer Research Institute, Vienna, Austria KATIA SCOTLANDI • IRCCS—Istituto Ortopedico Rizzoli, Experimental Oncology Laboratory, Bologna, Italy JEREMY M. SIMON • Department of Genetics, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; UNC Neuroscience Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA ELIZABETH STEWART • St. Jude Children’s Research Hospital, Memphis, TN, USA DIDIER SURDEZ • INSERM U830, E´quipe Labellise´e LNCC, Genetics and Biology of Pediatric Cancers, PSL Research University, SIREDO Oncology Centre, Institut Curie Research Centre, Paris, France MIWA TANAKA • Division of Carcinogenesis, The Cancer Institute, Japanese Foundation for Cancer Research, Tokyo, Japan OSCAR M. TIRADO • Sarcoma Research Group, Oncobell Program, Bellvitge Biomedical Research Institute (IDIBELL), L’Hospitalet de Llobregat, Barcelona, Spain; CIBERONC, Carlos III Institute of Health (ISCIII), Madrid, Spain MARCEL TRAUTMANN • Division of Translational Pathology, Gerhard-Domagk-Institute of Pathology, Mu¨nster, Germany ANDREI ZINOVYEV • Institut Curie, PSL Research University, Paris, France; INSERM U900, Paris, France; CBIO-Centre for Computational Biology, Mines ParisTech, PSL Research University, Paris, France

Part I Molecular Biology Techniques Applied To Ewing Sarcoma

Chapter 1 Germline Variation and Somatic Alterations in Ewing Sarcoma Mitchell J. Machiela and Thomas G. P. Gru¨newald Abstract Ewing sarcoma (EwS) is a rare bone or soft tissue tumor that occurs early in life and as such genetic variation is a major contributor to EwS risk. To date, genetic investigations have identified key somatic mutations and germline variants of importance for EwS risk. While substantial progress is being made in uncovering the genetic etiology of EwS, considerable gaps in knowledge remain. Herein, we provide a summary of methodological approaches for future genomic investigations of EwS. We anticipate this recommended analytical framework for germline and somatic investigations, along with genomic data from growing EwS case series, will aid in accelerating new genomic discoveries in EwS and expand knowledge of the genetic architecture of EwS. Key words Ewing sarcoma, EWSR1-FLI1, Genetics, Germline variants, SNPs, Somatic mutations, Translocation

1

Introduction Ewing sarcoma (EwS) is a tumor of childhood, adolescence, and adulthood. As such, relatively few environmental exposures have accumulated during the life course to substantially affect risk of EwS and, to date, no exogenous exposures have been robustly associated with EwS risk [1, 2]. Rather, risk of EwS appears to be largely driven by genetic determinants [3]. Studies of germline genetic variation and acquired somatic alterations in EwS cases offer tremendous potential to better understand the etiologic mechanisms that lead to EwS development. The aim of this chapter is to provide a framework of common methods and approaches for performing genomic investigations of EwS. Perhaps the largest motivator of genetic investigations of EwS is to find novel insights into EwS etiology. Whether these investigations identify inherited variants associated with EwS susceptibility or highlight common sets of acquired mutations shared across many EwS cases, such studies provide new clues into the underlying

Florencia Cidre-Aranaz and Thomas G. P. Gru¨newald (eds.), Ewing Sarcoma: Methods and Protocols, Methods in Molecular Biology, vol. 2226, https://doi.org/10.1007/978-1-0716-1020-6_1, © Springer Science+Business Media, LLC, part of Springer Nature 2021

3

4

Mitchell J. Machiela and Thomas G. P. Gru¨newald

genetic architecture of EwS. These genetic discoveries can provide important leads into promising research directions as well as accelerate advances into understanding the core regulatory networks key to EwS growth and survival. In addition, expanded knowledge of EwS genetic susceptibility can lead to improved genetic risk stratification and aid in identifying individuals or families at elevated risk of EwS for targeted screening or preventative measures. Ample evidence suggests a genetic component to EwS. Well before modern genomic analyses of EwS, observational studies suggested a disparity in EwS incidence by ancestry in which higher rates of EwS were observed in populations of European ancestry [4, 5]. These differences in EwS incidence by ancestry are observed presently [6] with rates of EwS higher in Europeans (1.5 cases per million individuals) as compared to Asians (0.8 cases per million individuals) and Africans (0.2 cases per million individuals). Remarkably, African Americans retain the relatively low incidence of EwS despite being exposed to many of the same environmental factors as Americans of European ancestry. Additionally, there have been incidental reports of EwS clustering in families, suggesting rare instances of familial aggregation of EwS, although evidence is limited [7]. Interestingly, EwS does not appear to be an outcome of well-described inherited cancer predisposition syndromes [8], although further studies are needed to better investigate potential relationships. The most well-characterized genomic component of EwS has been the characterization of pathognomonic somatic gene fusions seen in EwS cases. Evidence for recurrent EwS fusions between chromosome 11 and 22 were first reported in the 1988 [9] followed by evidence of gene fusions in 1992 by Delattre et al. [10]. Since these initial discoveries, chromosomal translocations involving a member of the FET gene family fused with an ETS family transcription factor have been demonstrated to be potent transcriptional dysregulators essential for the transformation of normal cells into EwS [11]. The most common of these fusions are EWSR1-FLI1 translocations seen in approximately 85% of EwS cases. These fusions typically result in an oncoprotein with a DNA-binding domain of the FLI1 component combined with a strong transcriptional enhancer. Apart from these fusions, EwS tumors are relatively mutationally quiet and characterized by very few (mostly nonrecurrent) other mutations. Examples of identified recurrent somatic mutations include mutations in STAG2, TP53, and CDKN2A at frequencies of around 20%, 5%, and 3%, respectively [3]. Inherited genetic variation has also been identified to be an important contributor to EwS susceptibility. Recent genome-wide association studies have identified common germline variation associated with EwS susceptibility [12, 13]. Interestingly, these EwS susceptibility loci are enriched near polymorphic GGAA

Methods in Ewing Sarcoma Genetics

5

microsatellite repeat sequences known to play an important role in EWSR1-FLI1 binding. As an example, a variant in the chromosome 10q21.3 region, rs79965208, associated with EwS susceptibility was found to disrupt a GGAA microsatellite sequence, thereby disrupting EWSR1-FLI1 binding to the region and impacting expression of EGR2, a gene with impacts on proliferation and clonogenicity in EwS cells [14]. Additionally, germline sequencing studies suggest a role of rare inherited inactivating mutations in DNA damage repair genes in approximately 13% of tumors, for example BRCA1 mutations, although their pathogenic roles remain unclear [15]. Substantial progress in identifying genetic determinants of EwS has been made in the past decade. Despite recent advances, the genetic etiology of EwS remains poorly characterized. As series of EwS cases continue to expand and new genomic approaches are developed, our understanding of the genetic etiology of EwS is anticipated to expand. This chapter outlines key methods and approaches for future genomic investigations.

2

Identifying a DNA Source for Investigation A key requirement for genetic studies of EwS is genomic material for investigation. While studies on both DNA and RNA provide insights into EwS etiology, this chapter will focus on current analysis approaches for studying variation in DNA important for EwS risk. Studies of DNA can generally be divided into two main categories: (1) studies on inherited germline genetic variation and (2) studies on acquired somatic variation. This chapter will expand on methods essential for both germline and somatic DNA analyses as they relate to EwS susceptibility. The genomic question of interest often dictates the sample source for DNA collection. For example, studies of somatic mutational profiles of EwS tumors will require DNA derived from excised primary tumor specimens or metastases. Current approaches enable DNA extraction from both fresh frozen tumor tissue as well as archival tumor tissue stored in formalin fixed paraffin embedded (FFPE) blocks. While fresh frozen tissue is preferred as DNA yield and quality are superior, many common genomic applications can be modified to include input DNA from FFPE tissue. Knowing which genotyping or sequencing technologies will be used in a genomic investigation will dictate whether DNA from fresh frozen or FFPE tissue is appropriate. For example, some sequencing approaches may require additional quality control filters or expanded purification or amplification steps before FFPE derived DNA can be used as an input.

6

Mitchell J. Machiela and Thomas G. P. Gru¨newald

Germline genetic investigations of EwS generally focus on nontumor tissue. Typical sources of DNA include whole blood, peripheral blood mononuclear cells, buccal epithelial cells, hair follicles, or bone marrow, as these tissues generally reflect germline DNA as inherited from parents. As EwS tumors are mutationally quiet, some studies have used tumor-derived DNA as a substitute for germline DNA [12]. While such studies have produced important new advances with respect to inherited variation important for EwS susceptibility, care should be taken to ensure genomic findings of interest reflect germline variation and not variation introduced by somatic mutations.

3

Considerations When Selecting EwS Cases and Controls One of the largest hurdles to investigating the genetic risk factors for EwS is the limited availability of EwS cases. As EwS is a rare tumor, acquiring large and well-powered series of EwS cases can be challenging. Most current genomic investigations of EwS are on only a few hundred EwS cases at most. While these sample sizes are large for a rare tumor, limitations in sample size impact power to detect key genetic relationships important to EwS etiology. The development of international EwS collaborations and consortia as well as sharing genomic data produced by these large efforts is vital for amassing large series of rare EwS cases for genomic investigation. Selecting the type of EwS cases to obtain genetic material from is an important consideration in the design of genetic studies. As EwS is a tumor predominantly observed in individuals of European ancestry [16], most genetic studies of EwS focus on individuals with a European genetic background. The enrichment of EwS among Europeans, however, does not preclude EwS studies in other ancestries. For example, admixture mapping studies of EwS cases with admixed ancestral backgrounds could provide unique opportunities for localizing European risk haplotypes in a background of non-European ancestry. Identifying a comparable control population is also essential for getting unbiased estimates for genetic effects. Researchers should ensure control series appropriately match the ancestral population from which the EwS cases arose. Many centers enrolling EwS cases fail to collect appropriate controls, often requiring controls to be obtained from independent series of cancer-free individuals. Furthermore, large age-matched populations of genotyped or sequenced disease-free controls often do not exist for the age range of individuals commonly affected by EwS. To circumvent the lack of collected controls for EwS genomic investigations, many researchers use cancer-free controls as a genomic comparison for EwS cases. Principal component matching can be used to ensure

Methods in Ewing Sarcoma Genetics

7

EwS cases are genetically well-matched to controls [13]; however, researchers exercising this approach for genomic analysis should be cautious of any potential survival biases that may be introduced from using adult controls for a cancer that predominantly occurs in childhood, adolescence and early adulthood. A final design consideration is the type of EwS cases included in the analysis. While EwS is well-defined by hallmark fusions, heterogeneity still exists with respect to age of diagnosis, tumor site, disease stage, and histological response to neoadjuvant therapy [3]. Investigators should be attentive to these individual and tumor characteristics when selecting cases for inclusion in genomic investigations. Imbalances in inclusion of certain subsets of EwS cases could result in findings that are not generalizable to all EwS cases. Alternatively, focusing on one particular subset of EwS cases could reduce heterogeneity in the analysis and aid in identifying novel genetic contributors to risk with application to a select subset of EwS cases.

4

Selection of Optimal Genomic Technologies The analytical question of interest also dictates which genomic technologies are well-suited to address the question. Often the choice of genomic technology is a balance between cost, sample size, and the detail or amount of genomic data produced. Two general approaches are used to produce genetic data: (1) arraybased approaches and (2) sequencing-based approaches; although other approaches also exist (e.g., TaqMan genotyping), they will not be the focus of discussion. Studies of germline variation and EwS can use both array-based as well as sequencing-based approaches. In general, studies interested in systematically scanning for associations with low-frequency and common variation use genotyping arrays. Currently available genotyping arrays include several hundred thousand to a few million markers on a commercially available platform and when intensity data is analyzed produce genotypes for the germline variants on the array. There are several companies that offer these arrays and different arrays can be chosen based on focus (e.g., oncologyrelated arrays, multiethnic arrays). These arrays are generally designed to have a backbone of variants that tag common and low-frequency variation across the genome. While the array only includes a small subset of all germline variation, the design enables for identification of regions or specific haplotypes that may be associated with EwS risk. In most cases, the tagging variants themselves are not the functional variants, but are highly correlated with germline variation that may be important for disease risk. At present, genotyping arrays are typically much cheaper than exome-wide

8

Mitchell J. Machiela and Thomas G. P. Gru¨newald

or genome-wide sequencing approaches, although improvements in next-generation sequencing approaches and methods may soon decrease these cost differences. Genotyping arrays perform well for identifying associations with common or low-frequency germline variation, but for germline investigations focused on rare variants, sequencing approaches are generally best. While Sanger sequencing [17] with chain terminating nucleotides still has application for regional sequencing experiments, most contemporaneous sequencing studies use massively parallel, short-read next-generation sequencing (NGS) approaches. There are several commercial vendors offering NGS products ranging in scope from single genes or regions of interest to exome-wide or genome-wide applications. Most NGS approaches require the construction of a library of fragmented DNA with adaptors to first be created followed by sequence-bysynthesis on a specially designed slide or flow cell within the sequencer. The sequencers produce large data files with sequence reads that are later aligned to a reference genome and eventually called for germline variants. Recently, long-read sequencing methods, in comparison to the short-read sequencing mentioned above, have also become widely available. These long-read sequencing approaches enable researchers to get better resolution of regional haplotype structure as well as improved sequencing of regions that are more difficult to sequencing by short-read sequencing (e.g., GGAA microsatellite regions that are of great importance to EwS pathogenesis [14, 18]). Costs for NGS is a function of the number of samples to sequence, the sequencing region(s) of interest and the target read depth. NGS is typically more expensive to perform than array-based genotyping. Similar to investigations of rare germline variants, investigations of somatic mutations in EwS tumor tissue almost exclusively use NGS approaches. Regions of interest for genomic investigation are targeted for sequencing with studies ranging in scope from single genes to exome-wide or genome-wide investigations. Genotyping arrays are not suited to studying somatic variants as variation included on these arrays are germline variants and unable to detect somatic mutations. Recent expansion of methods that use raw genotyping array intensity data, however, have been fruitful in identifying large somatic structural copy number and copy neutral variation in a variety of tissue types [19, 20].

5

Key Considerations in EwS Germline Association Analyses Equally important to generating genomic data is correctly analyzing this data. EwS germline association studies need to first ensure that genotype data is of sufficient quality before proceeding. A good rule of thumb is to run duplicate samples for a small

Methods in Ewing Sarcoma Genetics

9

percentage of study participants (e.g., a mixture of 5% cases and controls) to ensure genotyping concordance rates in excess of 99% between samples. Each sample and genotyped variant should also be checked for high genotype completion rates (generally >95%). Other recommended sample quality checks include checking for sex concordance, unexpected duplicates or related samples as well as samples that have heterozygosity rates that substantially deviate from expected. Most studies also perform Hardy–Weinberg equilibrium tests for each variant to ensure genotypes among controls do not substantially deviate from expected population proportions. These data filtering steps are easy to perform using genomics software such as in PLINK [21]. Genotype imputation is common practice to fill in missing genotypes as well as infer genotypes of variants that are not directly genotyped or sequenced. As detailed above, only high-quality genotypes and samples are used as input for imputation. When performing imputation, sample genotypes are first phased to determine underlying haplotypes and then a reference panel of similar ancestry to the study population is used to fill in missing genotypes. A popular phasing program is EAGLE2 [22] and a popular imputation program is minimac [23], both of which can be implemented using an online pipeline through the Michigan Imputation Server [23]. Common imputation reference panels include the TOPMed project [24], the Haplotype Reference Consortium (HRC) [25] and the 1000 Genomes Project [26]. As with genotyping data, imputed data is filtered to remove poorly imputed variants. Generally, imputed variants with imputation scores or R2 values in the range of 0.3–0.5 or below are removed from downstream analyses. Genotype imputation permits germline data generated from different platforms or labs to be merged for large-scale analysis. While merging is attractive for expanding EwS sample size, caution should be taken to ensure each merged set has EwS cases and a representative set of controls and additionally has performed uniform quality control and imputation procedures. In addition, confounding by ancestry (also referred to as population stratification) is a key factor of concern for many genetic association studies. A simple way to reduce the concern for confounding by ancestry is to restrict to individuals of one ancestry for study (e.g., European ancestry as EwS is overrepresented among Europeans); although, studies that include multiple ancestries or admixed individuals can have unique advantages but usually require advanced analytical approaches. EwS germline studies can use self-reported ancestry or compute ancestry from genomic information (e.g., using SNPWEIGHTS [27]). Confounding by ancestry can still occur even when restricting to one ancestral group due to subtle differences in allele frequencies among individuals within a population. One common adjustment approach is a dimension reduction method called principal component analysis (PCA) in which

10

Mitchell J. Machiela and Thomas G. P. Gru¨newald

principal components representing dimensions of ancestry are used to adjust out the effects of residual population stratification [28]. Commonly used software for PCA analysis includes EIGENSTRAT [28] and PLINK [21]. Many software suites have been developed for performing genetic association analyses. Popular tools include PLINK [21], SNPTEST [29], BOLT-LMM [30], and SAIGE [31]. These packages have varying options and approaches for performing tests of genetic association. After association tests are performed data is commonly plotted in Manhattan plots, plots of genomic position by -log10 p-value, to identify regions of association with EwS. The qqman R package [32] offers good functionality for generating Manhattan plots. Following association analyses, signals are further interrogated by disentangling LD patterns [33, 34] and looking at functional annotation [35, 36] to inform future functional studies.

6

Analysis Elements for NGS Projects The advent of NGS has resulted in expansive datasets of short-read sequencing data. Most modern sequencing platforms generate output files of many thousands of raw reads based on the collection of data during the sequencing process (e.g., Illumina HiSeq 4000 or NovaSeq). Initial quality control steps are first performed on these raw reads to remove poor quality reads as well as unnecessary adaptor or barcoding data. A key next step is the alignment of this short-read data to locations in the genome that these reads most optimally map to; although, not all reads will align to a location in the human genome. Current algorithms and alignment software have made substantial progress in speeding up genomic alignment of short read sequencing data. A common sequencing aligner is the Burrows–Wheeler Aligner available through the BWA-MEM software package [37]. For some EwS sequencing projects, more advanced alignment methods such as local realignment and deduplication of reads may be critical for generating high-quality sequencing alignments. After sequencing alignment is complete, variant calling can be performed. There are typically two different objectives of variant calling: (1) detecting germline variation or (2) detecting somatic variation. The type of variant calling will depend on the design goal and area of interest of your study. Germline variant calling is most commonly used in studies of EwS susceptibility. Common germline callers include the Genome Analysis Toolkit (GATK) HaplotypeCaller [38], Free Bayes [39], and Strelka2 [40]. Newer, more sensitive approaches to calling germline variants also incorporate joint calling of multiple aligners into consensus calls with higher accuracy. In addition to standard germline callers, callers specialized

Methods in Ewing Sarcoma Genetics

11

for detecting larger forms of structural variation are also coming online. While many of these callers are still in their infancy, platforms such as Lumpy [41], SnowmanSV, Manta [42], and SvABA [43] are helpful for detecting structural genetic variation. Somatic variant calling is commonly used to look for mutations that arise in EwS tumor tissue. Generally, somatic variant callers require a study design of matched tumor and normal samples from the same individuals or a panel of “normal” samples for differentiating somatic variants from inherited germline variation. Mutect2 [44] and Strelka2 [40] are two widely used somatic variant callers. Detection of somatic variants by joint calling of germline callers with a focus on detected variants with variant allele frequencies that substantially deviate from 50:50 Mendelian proportions is also possible in instances where no germline DNA is available for the EwS patient. Likewise, germline structural variant callers can also be used to call somatic structural variation; but as with detection of germline structural variants, calling somatic structural variants from short read sequencing data can be challenging. Linking detected genetic variation to EwS risk or progression is the primary aim of most EwS sequencing projects. Many of the analytic approaches mentioned above for array-based association studies also have relevance for sequencing-based association studies. A unique aspect of sequencing studies is that in addition to common variation, sequencing studies have the ability to detect low-frequency and rare variants not tagged well by array-based studies. These rare variants present challenges for statistical analysis as it is unlikely many EwS cases harbor exactly the same rare variant. Rather, it may be the case that several EwS cases carry different rare variants that all happen to cluster in a single gene, suggesting an important function of the gene for EwS. To investigate such instances of rare variant clustering specialized association tests have been developed that focus on aggregations of rare variants. An early developed rare variant aggregation test is SKAT [45], that allows for defined aggregations of rare variants while efficiently adjusting for other covariates. Other flavors of SKAT (e.g., SKATO [46]) as well as other extensions to burden tests have also become available for testing aggregations of rare variants from sequencing data [47]. A common goal of both germline and somatic EwS investigations is classifying associated variants based on predicted functional consequence. Most currently available variant annotation tools focus on variants within the coding region of genes as changes in nucleotide sequence in these regions are expected to have the greatest impact on protein function and subsequently EwS risk. Common variant annotation tools include COSMIC [48], ClinVar [49], CADD [50], REVEL [51], and Meta-SVM [52]. While these tools are helpful in predicting impact of protein coding changes or identifying associations with other diseases, the level of support for

12

Mitchell J. Machiela and Thomas G. P. Gru¨newald

predictions vary considerably and as such these tools should only be used to provide insights into potential relationships between genetic variants and disease risk. In most cases, further functional investigations and more refined association studies are needed to conclusively link genetic variants to altered EwS risk or response.

7

Final Considerations for EwS Genetic Investigations This chapter on EwS genetic methods and protocols is intended to serve as a starting point for genomic investigations of EwS. Depending on the nature of specific EwS genomic investigations, further methodologic needs or more advanced analytic approaches may be required. It is recommended to collaborate with a diverse team of collaborators on EwS genomic investigations with individual expertise in areas such as biology, epidemiology, genetics, bioinformatics, biostatistics, and oncology. Additionally, while many parts in the preceding sections made a distinction between somatic and germline investigations, recent EwS genomic investigations indicate germline–somatic interactions are likely key contributors to the genetic etiology of EwS [13, 14, 53]. As such, integrative studies that combine data on germline variation and somatic mutations along with rich genomic annotation data from gene expression, chromatin organization, DNA methylation and ChIP-seq are likely to provide important new insights for expanding knowledge of EwS risk.

References 1. Moore LE, Gold L, Stewart PA et al (2005) Parental occupational exposures and Ewing’s sarcoma. Int J Cancer 114:472–478 2. Valery PC, McWhirter W, Sleigh A et al (2002) Farm exposures, parental occupation, and risk of Ewing’s sarcoma in Australia: a national case-control study. Cancer Causes Control CCC 13:263–270 3. Gru¨newald TGP, Cidre-Aranaz F, Surdez D et al (2018) Ewing sarcoma. Nat Rev Dis Primer 4:5 4. Fraumeni JF, Glass AG (1970) Rarity of Ewing’s sarcoma among U.S. Negro children. Lancet Lond Engl 1:366–367 5. Jensen RD, Drake RM (1970) Rarity of Ewing’s tumour in Negroes. Lancet Lond Engl 1:777 6. Jawad MU, Cheung MC, Min ES et al (2009) Ewing sarcoma demonstrates racial disparities in incidence-related and sex-related differences in outcome: an analysis of 1631 cases from the SEER database, 1973–2005. Cancer 115:3526–3536

7. Joyce MJ, Harmon DC, Mankin HJ et al (1984) Ewing’s sarcoma in female siblings. A clinical report and review of the literature. Cancer 53:1959–1962 8. Rahman N (2014) Realizing the promise of cancer predisposition genes. Nature 505:302–308 9. Turc-Carel C, Aurias A, Mugneret F et al (1988) Chromosomes in Ewing’s sarcoma. I. An evaluation of 85 cases of remarkable consistency of t(11;22)(q24;q12). Cancer Genet Cytogenet 32:229–238 10. Delattre O, Zucman J, Plougastel B et al (1992) Gene fusion with an ETS DNA-binding domain caused by chromosome translocation in human tumours. Nature 359:162–165 11. Sankar S, Lessnick SL (2011) Promiscuous partnerships in Ewing’s sarcoma. Cancer Genet 204:351–365 12. Postel-Vinay S, Ve´ron AS, Tirode F et al (2012) Common variants near TARDBP and EGR2

Methods in Ewing Sarcoma Genetics are associated with susceptibility to Ewing sarcoma. Nat Genet 44:323–327 13. Machiela MJ, Gru¨newald TGP, Surdez D et al (2018) Genome-wide association study identifies multiple new loci associated with Ewing sarcoma susceptibility. Nat Commun 9:3184 14. Gru¨newald TGP, Bernard V, GilardiHebenstreit P et al (2015) Chimeric EWSR1FLI1 regulates the Ewing sarcoma susceptibility gene EGR2 via a GGAA microsatellite. Nat Genet 47:1073–1078 15. Brohl AS, Patidar R, Turner CE et al (2017) Frequent inactivating germline mutations in DNA repair genes in patients with Ewing sarcoma. Genet Med Off J Am Coll Med Genet 19:955–958 16. Worch J, Cyrus J, Goldsby R et al (2011) Racial differences in the incidence of mesenchymal tumors associated with EWSR1 translocation. Cancer Epidemiol Biomarkers Prev 20:449–453 17. Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A 74:5463–5467 18. Gangwal K, Sankar S, Hollenhorst PC et al (2008) Microsatellites as EWS/FLI response elements in Ewing’s sarcoma. Proc Natl Acad Sci U S A 105:10149–10154 19. Loh P-R, Genovese G, Handsaker RE et al (2018) Insights into clonal haematopoiesis from 8,342 mosaic chromosomal alterations. Nature 559:350–355 20. Machiela MJ, Zhou W, Sampson JN et al (2015) Characterization of large structural genetic mosaicism in human autosomes. Am J Hum Genet 96:487–497 21. Purcell S, Neale B, Todd-Brown K et al (2007) PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 81:559–575 22. Loh P-R, Danecek P, Palamara PF et al (2016) Reference-based phasing using the Haplotype Reference Consortium panel. Nat Genet 48:1443–1448 23. Das S, Forer L, Scho¨nherr S et al (2016) Nextgeneration genotype imputation service and methods. Nat Genet 48:1284–1287 24. Brody JA, Morrison AC, Bis JC et al (2017) Analysis commons, a team approach to discovery in a big-data environment for genetic epidemiology. Nat Genet 49:1560–1563 25. McCarthy S, Das S, Kretzschmar W et al (2016) A reference panel of 64,976 haplotypes for genotype imputation. Nat Genet 48:1279–1283

13

26. Auton A, Abecasis GR, Altshuler DM et al (2015) A global reference for human genetic variation. Nature 526:68–74 27. Chen C-Y, Pollack S, Hunter DJ et al (2013) Improved ancestry inference using weights from external reference panels. Bioinforma Oxf Engl 29:1399–1406 28. Price AL, Patterson NJ, Plenge RM et al (2006) Principal components analysis corrects for stratification in genome-wide association studies. Nat Genet 38:904–909 29. Marchini J, Howie B, Myers S et al (2007) A new multipoint method for genome-wide association studies by imputation of genotypes. Nat Genet 39:906–913 30. Loh P-R, Tucker G, Bulik-Sullivan BK et al (2015) Efficient Bayesian mixed-model analysis increases association power in large cohorts. Nat Genet 47:284–290 31. Zhou W, Nielsen JB, Fritsche LG et al (2018) Efficiently controlling for case-control imbalance and sample relatedness in large-scale genetic association studies. Nat Genet 50:1335–1341 32. Turner S (2018) qqman: an R package for visualizing GWAS results using Q-Q and manhattan plots. In: J. Open Source Softw. https:// doi.org/10.1101/005165. Accessed 27 Nov 2019 33. Machiela MJ, Chanock SJ (2015) LDlink: a web-based application for exploring population-specific haplotype structure and linking correlated alleles of possible functional variants. Bioinforma Oxf Engl 31:3555–3557 34. Machiela MJ, Chanock SJ (2018) LDassoc: an online tool for interactively exploring genomewide association study results and prioritizing variants for functional investigation. Bioinforma Oxf Engl 34:887–889 35. Ward LD, Kellis M (2012) HaploReg: a resource for exploring chromatin states, conservation, and regulatory motif alterations within sets of genetically linked variants. Nucleic Acids Res 40:D930–D934 36. Boyle AP, Hong EL, Hariharan M et al (2012) Annotation of functional variation in personal genomes using RegulomeDB. Genome Res 22:1790–1797 37. Li H, Durbin R (2009) Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinforma Oxf Engl 25:1754–1760 38. DePristo MA, Banks E, Poplin R et al (2011) A framework for variation discovery and genotyping using next-generation DNA sequencing data. Nat Genet 43:491–498

14

Mitchell J. Machiela and Thomas G. P. Gru¨newald

39. Garrison E, Marth G (2012) Haplotype-based variant detection from short-read sequencing. ArXiv12073907 Q-Bio 40. Kim S, Scheffler K, Halpern AL et al (2018) Strelka2: fast and accurate calling of germline and somatic variants. Nat Methods 15:591–594 41. Layer RM, Chiang C, Quinlan AR, Hall IM (2014) LUMPY: a probabilistic framework for structural variant discovery. Genome Biol 15: R84 42. Chen X, Schulz-Trieglaff O, Shaw R et al (2016) Manta: rapid detection of structural variants and indels for germline and cancer sequencing applications. Bioinformatics 32:1220–1222 43. Wala JA, Bandopadhayay P, Greenwald N et al (2018) SvABA: genome-wide detection of structural variants and indels by local assembly. Genome Res 28(4):581–591. https://doi. org/10.1101/gr.221028.117 44. Cibulskis K, Lawrence MS, Carter SL et al (2013) Sensitive detection of somatic point mutations in impure and heterogeneous cancer samples. Nat Biotechnol 31:213–219 45. Wu MC, Lee S, Cai T et al (2011) Rare-variant association testing for sequencing data with the sequence kernel association test. Am J Hum Genet 89:82–93

46. Lee S, Emond MJ, Bamshad MJ et al (2012) Optimal unified approach for rare-variant association testing with application to small-sample case-control whole-exome sequencing studies. Am J Hum Genet 91:224–237 47. Lee S, Abecasis GR, Boehnke M, Lin X (2014) Rare-variant association analysis: study designs and statistical tests. Am J Hum Genet 95:5–23 48. Tate JG, Bamford S, Jubb HC et al (2019) COSMIC: the catalogue of somatic mutations in cancer. Nucleic Acids Res 47:D941–D947 49. Landrum MJ, Lee JM, Riley GR et al (2014) ClinVar: public archive of relationships among sequence variation and human phenotype. Nucleic Acids Res 42:D980–D985 50. Rentzsch P, Witten D, Cooper GM et al (2019) CADD: predicting the deleteriousness of variants throughout the human genome. Nucleic Acids Res 47:D886–D894 51. Ioannidis NM, Rothstein JH, Pejaver V et al (2016) REVEL: an ensemble method for predicting the pathogenicity of rare missense variants. Am J Hum Genet 99:877–885 52. Kim S, Jhong J-H, Lee J, Koo J-Y (2017) Meta-analytic support vector machine for integrating multiple omics data. BioData Min 10 53. Musa J, Cidre-Aranaz F, Aynaud M-M et al (2019) Cooperation of cancer drivers with regulatory germline variants shapes clinical outcomes. Nat Commun 10:1–10

Chapter 2 Western Blot Analysis in Ewing Sarcoma Aruna Marchetto and Laura Romero-Pe´rez Abstract Western blot is an experimental method used to analyze protein expression. In Ewing sarcoma, as in many other diseases, Western blot provides information about the level of protein expression in different cell conditions, in comparison with other tissues or upon induced molecular changes. Based on the specific pattern of protein expression of the tissue, as well as on the characteristics of the protein of interest, the antibodies and protocol of Western blot may be modified according to different specifications. Here we describe some of these peculiarities in frame of Ewing sarcoma field. Key words Western blot, Ewing sarcoma, Protein expression

1

Introduction Western blot is a common molecular method used to detect the expression of one or more proteins of interest. This technique was described in 1981 by W. Neal Burnette [1], who named it based on? the similarity with previously described methods for the transference of DNA (Southern blot) or RNA (Northern blot) to a nitrocellulose membrane.

1.1 Western Blot, the Technique

In Ewing sarcoma, as in many other fields of biological and clinical research, Western blot is used to quantify the expression of specific proteins in different cell lines or even tissues, in an exploratory way or upon different experimental events. Examples of those experimental assays are: knocking out/down/in of certain genes; the induction of changes in a regulatory pathway; the inhibition/ induction of protein expression by a specific drug treatment; the induction of protein modifications such as phosphorylation/ dephosphorylation; or to detect target proteins obtained from previous assays such as immunoprecipitation. In summary, the protocol of a Western blot is based on an electrophoretic process by which the proteins contained in a buffered lysate are charged in a gel (sodium dodecyl sulfate-

Florencia Cidre-Aranaz and Thomas G. P. Gru¨newald (eds.), Ewing Sarcoma: Methods and Protocols, Methods in Molecular Biology, vol. 2226, https://doi.org/10.1007/978-1-0716-1020-6_2, © Springer Science+Business Media, LLC, part of Springer Nature 2021

15

16

Aruna Marchetto and Laura Romero-Pe´rez

polyacrylamide gel, SDS-PAGE, Note 1) and separate according to their mass and electric charge (see Note 2). The pattern of separated proteins is then transferred to a membrane (see Note 3) to be then blocked (see Note 4) and incubated with the specific antibody (polyclonal or monoclonal; Note 5) against the protein of interest. Both the blocking step and the use of specific antibodies aim at minimizing nonspecific binding to other proteins. Washing and incubation with a secondary antibody (reactive to the primary antibody host species) is required to amplify the signal through a conjugated enzyme such as horseradish peroxidase (HRP) or alkaline phosphatase (AP). The most commonly used techniques to visualize the results are developing films or transillumination. By using transillumination, fluorescent, colorimetric or more frequently, chemiluminescent reagents can be detected depending on the desired sensitivity (see Note 6). To analyze the results, the normalization of the expression of the protein among the different conditions is performed with the use of a loading control. It is based on the use of an antibody against a housekeeping protein, meaning a protein constitutively expressed in all cells of interest, which size should be different from the proteins that will be analyzed. In order to estimate the size of the bands (which is the same as to identify the detected proteins) a protein ladder must be charged in the gel. Color markers for certain sizes facilitate the identification of the protein of interest. In the recent years, single cell-resolution Western blotting has been described [2, 3], even at the subcellular level [4]. This application is especially useful when avoiding background signal of surrounding cells is required. However, to our knowledge, these interesting methods have not been applied to the field of Ewing sarcoma yet probably due to the particularity of this kind of tumor showing “small round cells” with a little proportion of cytoplasm. Although this may present some difficulties, Ewing sarcoma researchers are encouraged to set up the conditions of these new methods in this field. 1.2 Ewing Sarcoma Specifications

Ewing sarcoma is characterized by recurrent balanced chromosomal translocations between EWSR1 gene (22q12.2) and FLI1 (11q24.3) in around 85% of cases, but also ERG (10–15%), ETV1, ETV4, or FEV can fuse to EWSR1 with phenotypically and clinically similar pattern? [5]. When the most common translocation is produced, exons 7 or 10 of EWSR1 are fused to exon 6 (Type I), exon 5 (Type II), or exon 7 (Type IV) of FLI1, or exons 6, 7, or 9 of ERG. Here the epitope of EWSR1 becomes unrecognizable, as it maps between amino acids 2–19 at the N-terminus in humans. Other fusion points have been described to produce the same loss of recognition of EWSR1 [6]. This fact explains why most specific antibodies used

Western Blot Analysis in Ewing Sarcoma

17

Fig. 1 Western blot showing endogenous EWSR1-FLI1 () in Ewing sarcoma cell lines (Rh1, SK-ES-1, and SK-N-MC), which harbor a dox-inducible shRNA against EWSR1-FLI1 (+)

to identify Ewing sarcoma cells from a total cell extract are specific for either FLI1 or ERG (as the most common partners of EWSR1), and CD-99 (a cell surface glycoprotein highly expressed in Ewing sarcoma; [7]). Importantly polyclonal antibodies against the transcription factor FLI1 may not only detect endogenous FLI1 but also FLI1 as part of the fusion protein. To ensure the detection of the fusion protein and not the endogenous FLI1, it is important to confirm the size of the obtained band and to use specific controls based on the presence or absence of EWS-FLI1 (Fig. 1) [8]. Figure 1 shows that upon the doxycycline induction of a shRNA against EWSR1FLI1 in Ewing sarcoma cell lines, the band of approx. 68 kDa corresponding to the fusion protein expected size [9] is reduced or completely lost. Folpe et al. described that among normal tissues only endothelial cells and small lymphocytes express FLI1 [10]. This is nowadays debatable, since it is registered in The Human Protein Atlas [11, 12] that cells of the brain, lung, breast and female reproductive system or renal, urinary, and male reproductive system also express FLI1.

2

Materials Prepare all the solutions for the Western blot analysis with distilled water and keep all the buffers at room temperature (RT) except for the RIPA buffer and the 1 Blotting buffer (4  C). Never use sodium azide neither for the buffer nor for the antibody solutions as it is known to interfere with the HRP-coupled secondary antibody.

2.1

Reagents

1. RIPA Buffer: Mix 150 mM NaCl (3.5 g), 0.5% sodium deoxycholate (5 g), 50 mM Tris–HCl, pH 8, 1% Triton X-100 (4 mL), 0.1% SDS (1 g) to a total volume of 1 L H2O and adjust the pH 8. Add every time fresh to the sample buffer: 1 mM Na3VO4 and 100 μL/mL Protease Cocktail inhibitor. Store at 4  C.

18

Aruna Marchetto and Laura Romero-Pe´rez

2. Bradford reagent: Dilute in a ratio 1:5 the Bradford reagent with H2O. 3. BSA standard: Dilute Bovine Serum Albumin (BSA) (0.125 μg, 0.25 μg, 0.5 μg, 0.75 μg, 1 μg, and 2 μg) with H2O to generate the standard curve to determine protein concentration afterward. 4. 10 TBS buffer, pH 7.3: Mix 24 g Tris base, 88 g NaCl and fill up to 1 L with H2O. Adjust the pH 7.3. Store at RT. 5. 1 TBST (0.1%): 100 mL 10 TBS buffer, 900 mL H2O and 1 mL Tween-20 Store at RT. 6. 10 Running/Blotting buffer (R/B): Mix 30 g Tris base and 144 g glycine to 1 L H2O. Store at RT. 7. 1 Running buffer: Mix 100 mL 10 R/B buffer with 10% SDS (100 g) and fill up with 900 mL H2O. Store at RT. 8. 1 Blotting buffer: Mix 100 mL 10 R/B buffer with 200 mL Methanol and fill up with 700 mL H2O. Store at 4  C. 9. 4 Loading dye: Mix 312 mM Tris–HCl, pH 6.8, 10% SDS (100 mg), 50% (500 μL) glycerol and add a bit of Bromophenol Blue for the color. Prepare aliquots and store at 20  C. Add fresh 250 mM DTT to 1 mL loading dye and keep up to 3 months in the fridge. 10. Blocking buffer: Mix 5% non-fat milk (2.5 g) or 5% BSA (2.5 g) with 20 mL of 1 TBST buffer, mix and fill up to 50 mL. Prepare fresh every time. 11. Ammonium persulfate (APS): Dissolve 10% APS (100 mg) in 1 mL of H2O. Prepare aliquots and store at 20  C. 12. Sodium dodecyl sulfate (SDS): Dissolve 10% SDS (100 mg) in 1 mL H2O. Store at RT. 13. N, N, N, N0 -Tetramethyl-ethylenediamine (TEMED): Store at RT. 14. 30% bisacrylamide (37.5:1): Store at 4  C. 15. 1.5 M Tris–HCl, pH 8.8: Dissolve 45.4 g Tris–HCl in 100 mL H2O and fill up to 250 mL. Adjust at pH 8.8 with HCl and store at RT. 16. 1 M Tris–HCl, pH 6.8: Dissolve 30.3 g Tris–HCl in 100 mL H2O and fill up to 250 mL. Adjust at pH 6.8 with HCl and store at RT. 17. Mouse monoclonal antibody against FLI1 (MRQ-1, Cell Marque). Observed band at ~70 kDa (primary antibody, Note 7). 18. Mouse monoclonal antibody against GAPDH (sc-32233, Santa Cruz). Observed band at ~37 kDa (primary antibody). 19. Polyclonal anti-mouse IgG (H + L) coupled to HRP (secondary antibody, Note 8).

Western Blot Analysis in Ewing Sarcoma

2.2

Equipment

19

1. Spectrophotometer for measurement of protein absorption. 2. SDS–polyacrylamide gel: glass plates (15 mm), combs (10, 15 chambers), prestained molecular weight standards. 3. Immunoblotting: Nitrocellulose membranes, blotting paper.

3

Methods All steps should be performed at room temperature, if not otherwise specified. This protocol is mainly focused on protein detection from Ewing sarcoma cell lines.

3.1 Sample Preparation

1. Prepare 100 μL RIPA buffer supplemented with Protease cocktail inhibitor and Phosphatase inhibitor (Na3VO4) per 6-well sample and lyse the cells (see Note 9) on ice with a cell scraper (Fig. 2a). 2. Transfer the cell lysate into a reaction tube and incubate for 30 min on ice while shaking (Fig. 2b). 3. Spin down at 11,000  g for 15 min and transfer the supernatant into a new reaction tube. For long-term storage, keep lysates at 80  C. 4. Measure protein concentration (see Note 10), add 1 Loading dye supplemented with 250 mM DTT (see Note 11). 5. Incubate at 95  C for 10 min to disrupt S–S boundaries (see Note 12, Fig. 2c). 6. Shortly spin down the lysate to bring down the condensate. Cool down on ice and keep on ice or freeze.

3.2 SDS–PAGE Gel Electrophoresis

1. Resolving gel (7%): Add 3.7 mL of H2O and 2 mL of 1.5 M Tris–HCl, pH 8.8 with 1.8 mL of 30% bisacrylamide in a 50 mL falcon. Mix thoroughly. Add 75 μL of 10% SDS, 40 μL of 10% APS and 10 μL TEMED to the previous solution and mix well (see Note 13). 2. Cast the gel within the glass plates with 1.5 mm thickness (10 cm  7.5 cm). To avoid bubbles overlay the surface of the resolving gel with isopropanol. Wait until the remaining solution in the falcon is solidified and proceed with the stacking gel preparation. 3. Stacking gel (3%): Add 3.5 mL of H2O and 860 μL of 1 M Tris–HCl, pH 6.8 with 500 μL of 30% bisacrylamide in a 5 mL falcon. Mix thoroughly. Add 47.5 μL 10% SDS, 10% APS and 10 μL TEMED to the previous solution and mix. 4. Remove the isopropanol by pouring the isopropanol from the glass plates into a paper towel, pour the stacking gel on top of

20

Aruna Marchetto and Laura Romero-Pe´rez

Fig. 2 Scheme of Western blot procedure. RT: room temperature, ON: overnight. (a) Cultured cells, (b) lysis, (c) denaturation, (d) gel electrophoresis, (e) protein transfer, (f) blocking, (g) primary antibody, (h) secondary antibody, (i) detection

the resolving gel and insert the corresponding 1.5 mm comb avoiding bubbles. 5. Insert the casted glass plates within the Western blot device and fill up with 1 Running buffer until marked in the device. 6. Load the prepared samples (10–20 μg protein concentration) and the protein ladder on the outer position and in the middle of the lanes in the corresponding pockets of the gel (see Note 14). 7. Electrophorese at 90 V until the dye front reaches the bottom of the gel (Fig. 2d). 3.3 Protein Transference

1. Take the glass plates with the gel out of the device and pry the glass plates with the help of a spatula. Remove the stacking gel from one side of the glass plate with a spatula and cut out the upper part of the stacking gel. Be always aware of the order in which the samples have been loaded for future reference (see Note 15).

Western Blot Analysis in Ewing Sarcoma

21

2. Cut the nitrocellulose membrane (see Note 16) and the blotting paper according to the size of the gel (10 cm  7.5 cm) and bath them in cold 1 Blotting buffer. 3. Prepare the wet-blot device (see Note 17) on a magnet stirrer and load half of the chamber with a frozen thermal pack and 1 Blotting buffer. 4. Load the elements in the following exact order while immersed in 1 Blotting buffer to avoid bubble formation (Fig. 2e): Cathode () sponge–two blotting papers–gel–nitrocellulose membrane–two blotting papers–sponge–anode (+). 5. Squeeze the air bubbles out of the sandwich with help of the spatula and insert it into the wet-blot device. 6. Let the proteins transfer for 110 min at 110 V while mixing on the magnetic stirrer (see Note 18). 3.4 Antibody Staining

1. Take out the sandwich from the wet-blot device and block the membrane by incubating for 1 h at RT with Blocking buffer (5% BSA dissolved in 1 TBST) while shaking (see Note 19, Fig. 2f). 2. Wash the membrane briefly with fresh 1 TBST buffer. 3. Cut the membrane according to the size of the proteins of interest and incubate each part with the corresponding primary antibody: anti-FLI1 (1:1000) in 5% BSA dissolved in 1 TBST and anti-GAPDH (1:2000) in 5% non-fat milk dissolved in 1 TBST overnight at 4  C while rotating (Fig. 2g). 4. Wash the nitrocellulose membranes three times with fresh 1 TBST for 10 min while shaking. 5. Secondarily incubate with anti-mouse IgG (H + L) HRP (1:3000) with 5% non-fat milk TBST for 1 h (see Note 20, Fig. 2h). 6. Wash the membranes three times with fresh 1 TBST for 10 min each while shaking. 7. Drain the wash buffer from the membrane prior to incubating the membrane with ECL (time: 1 min). The protein amount is detected via chemiluminescence as HRP cleaves luminol (Fig. 2i).

3.5

Analysis

The analysis should be performed by comparing the intensity of the band of the protein of interest with respect to the intensity of the band of the housekeeping protein. The ratio should be calculated for the same specimen and also by comparison with the ratio corresponding to the rest of conditions or specimens analyzed in the same assay. The intensity estimation can be performed with different software such as Image J (NIH) [13].

22

4

Aruna Marchetto and Laura Romero-Pe´rez

Notes 1. The percentage of acrylamide will depend on the size of the protein of interest. Normally, 4–7% is recommended for proteins of 100–300 kDa and up to 10–12.5% for proteins 70–80%) in the hematoxylin–eosin staining corresponding to each FFPE block (Fig. 2). 2. Localize this area in the FFPE block. 3. Isolate cylinders of tissue by puncturing with a 2-mm biopsy punch (Fig. 2). 4. Cut the tissue core into very small pieces with the scalpel, or triturate/homogenize, to facilitate the subsequent digestion with Proteinase-K required for DNA/RNA isolation. This is very useful when working with small biopsies from pediatric patients such as those with suspected EwS.

3.4.2 Deparaffination

Nucleic acids contained in FFPE samples are inevitably highly degraded due to the traditional method of fixation in formalin. The procedures of deparaffination and isolation of nucleic acids from FFPE samples should be the less aggressive possible to offer clean and of high-quality results.

DNA/RNA Isolation from Ewing Sarcoma FFPE Samples

33

Fig. 2 Tumor cell enrichment by localizing and puncturing the tumor area of an Ewing sarcoma (EwS) FFPE sample. Tumor area is marked in red and the percentage of tumor cells is annotated. The corresponding FFPE block has been punctured with a 2 mm biopsy punch in the selected area

1. Use mineral oil (from the FormaPure total kit; Note 18) to dissolve the paraffin from the tissue with the help of the temperature for a short period of time (80  C, 5 min; Note 19) according to the manufacturer’s instructions. 2. Add the lysis buffer, centrifuge and add Proteinase-k according to the manufacturer instructions. An interface will be formed and the mineral oil will stay in the upper part without contacting the sample. 3.4.3 DNA and RNA Isolation

Many kits are commercially available for the isolation of DNA and/or RNA from FFPE samples. Most of these kits include steps of centrifugation to make the sample be filtered through a column, retaining the nucleic acids in the filter and letting the washing solution pass to the lower part to be discarded. In this way, some of the nucleic acids remain retained in the filter even after the last elution step, obtaining less amount of nucleic acids than the one potentially contained in the sample. On the contrary, we propose to use isolation methods based on nucleic acids binding to magnetic beads (SPRI beads), like the FormaPure total kit. It has been optimized for using before downstream next generation sequencing (NGS) and genotyping assays, which are sometimes required for complex sarcoma cases. In

34

Laura Romero-Pe´rez and Thomas G. P. Gru¨newald

addition, this kit offers the possibility of isolating both DNA and RNA from the same FFPE sample, which is an advantage when the amount of tumor tissue is reduced. 1. For RNA isolation a binding solution is added to immobilize the nucleic acids to the surface of the magnetic beads. Contaminants are rinsed away using a simple washing procedure, DNA is removed from the sample, and RNA is again immobilized on the surface of the beads before eluting with water (see Note 20 for enzymatic treatment). 2. For DNA isolation, RNA is removed from the sample, and a binding solution is added to immobilize the DNA to the surface of the magnetic beads. Contaminants are rinsed away using a simple washing procedure, and the DNA is eluted with water (see Note 20). 3.4.4 Measurement of the Concentration and Quality of Nucleic Acids

Nanodrop is usually chosen as the fastest and easiest method to check at the same time the concentration and estimated quality of isolated DNA or RNA. We recommend to pay attention to both, 260/280 and 260/230 absorbance ratios to calculate an index of quality (which should be next to 1, and separately each of them should be close to 2), in addition to the curve pattern. However, this method is not the most precise and it could just help to decide if a sample should be discarded or not from the cohort. For more accurate nucleic acids concentration estimation, we recommend the Qubit fluorimeter rather than UV-absorbance quantification methods, being able to distinguish between DNA and RNA in the same sample. It can quantify the concentration from very low concentrated samples (10 pg/μl) to 1 μg/μl, but it is not able to detect contaminants as it can be revealed or estimated by the NanoDrop curve. Moreover, Qubit measurements require to perform a protocol consisting in preparing a working solution where to dilute each sample, as well as the nucleic acids standards to build a concentration curve. Sometimes, very low or highly concentrated samples may offer an out-of-range result, for which it will be necessary to repeat the process. Since Qubit provides very precise information about nucleic acids concentration, but it does not provide any information about the quality of the sample, we recommend to additionally use the Bioanalyzer for subsequent NGS analysis. This method, based in electrophoresis, may result expensive because of the use of especial DNA- or RNA-chips, reagents and device, but it is very informative. It provides information about sizing, quantitation, integrity and purity from DNA and RNA, expending minimal sample volumes. Information about the amount of DNA/RNA fragments and the average size is crucial when working with FFPE samples. For example, DV200 values (the percentage of fragments >200 nucleotides) are of especial interest for sequencing assays.

DNA/RNA Isolation from Ewing Sarcoma FFPE Samples

4

35

Notes 1. Methanol is a stabilizer that slows the process by which the polymerized PFA would precipitate in water. Buffers (usually phosphate buffers, but also magnesium carbonate, citrate or Tris) are added to formalin to maintain the pH close to 7 and to avoid the oxidation of formaldehyde to formic acid, which would produce pigment granules precipitation [6]. 2. The method of fresh tissue preservation is similar to the one usually used for organ transplantation. Preservation solutions have been classified as: intracellular fluid (ICF)-type solution with a high potassium–low sodium ratio; or extracellular fluid (ECF)-type preservation solution with a low-potassium–highsodium ratio, being this last the one recently demonstrated to improve the functionality of the organs. The most commonly used ECF-type solutions are PBS or DMEM, although it has been recently described another particular ECF-type preservation solution with better results in terms of viability of tissue stem cells for longer period of time [2]. 3. From the frequent type of samples received in the pathology routine, frozen samples are the most valuable sources for molecular assays. If these tissue samples cannot be processed immediately after the resection, they can be immersed in a tube with RNAlater and to store overnight at 2–8  C in an upright position, to ensure the conservancy of the RNA [7]. To do so, the size should not exceed 0.5  0.5 cm in height and width. After overnight incubation, samples can be stored in RNAlater at room temperature (RT), 4  C or 20  C, for some time, but it is much better to store directly in liquid nitrogen or 80  C for long term. 4. The use of cryopreservative solutions helps to avoid the damage of tissue due to ice crystal formation during freezing. Ice crystals would alter the morphology, damage cell membranes, and finally affect histologic evaluation. The appearance at microscope of a specimen with ice crystals will be like a perforated surface of tissue, losing the architecture of potentially informative areas. 5. When working with a frozen block, the side that will be first sectioned is the one touching the bottom of the cryomold, so the sample should be placed accordingly and properly orientated. While adding OCT, try to avoid the formation of air bubbles. 6. Instead of the only used of OCT, other methods include a previous step of fixation in 4% paraformaldehyde (PFA),

36

Laura Romero-Pe´rez and Thomas G. P. Gru¨newald

followed by cryoprotection in sucrose (usually 30% in phosphate buffer). This allows to use a slower freezing method but generates fixative cross-linking, making necessary to perform antigen retrieval before immunohistochemistry (IHC). 7. It is not recommended to use methanol, ethanol or acetone instead of isopentane, because when OCT contacts any of these reagents, it is not able to freeze down properly. 8. A cryostat is a special microtome maintaining constant temperature ( 10  C and below) during the whole process, so the freeze–thaw cycles should be taken into account, but the degradation is not as extreme as the one caused by freezing and thawing repeated times at RT. 9. OCT is a water-soluble mixture of glycols and resins, which does not leave any residues that may interfere during staining procedures. Usually, as no cross-linking is produced by any fixative, antigen retrieval is not required and epitopes are shown in an almost native state. 10. Even if liquid nitrogen seems to be the best option for snap freezing because it does not mix with the tissue, there are some specific problems related with the boiling effect in contact with the tissue at RT. This effect generates a vapor barrier next to the tissue surface inducing a slow penetration of cold, becoming the rate of freezing imprecise. This, together with the fact that biological tissues are poor thermal conductors, make possible that a thermal gradient can be established from the surface to the core of the tissue during the freezing process, especially if sample is not very thin. Thus, this method is only recommended to freeze biopsies or very small resection samples. 11. Although the method of surrounding the sample with powdered dry ice could be also an option, it is less recommended. Even to introduce the sample in a freezer is possible but any of these methods can freeze the tissue quickly enough. 12. For molecular tests liquid nitrogen snap-frozen samples can be stored in cryovials for long term. The only limitation is that in this case, the whole sample is thawed for DNA/RNA/protein isolation, so no different procedures or isolations will be performed in the future from the same sample. 13. When fixating a sample it is important to take into account the fixative solution chosen, the time of fixation and the thickness and properties of the tissue. Fixatives can be divided in four groups including aldehydes, oxidizing agents, alcohol-based and metallic group of fixatives [8]. Formalin, made with formaldehyde, can penetrate into the tissue at an average of 1 mm/ h [1], depending on the tissue type. Once inside the tissue, the reaction of cross-linking between formaldehyde and proteins

DNA/RNA Isolation from Ewing Sarcoma FFPE Samples

37

starts and triggers the formation of stable methylene bridges, which provides harshness to the tissue. 14. Usually, 24 h is enough for the formalin fixation reaction, depending on the thickness of the tissue and the temperature [8], but sometimes 36–48 h may be necessary, increasing the fragmentation of the nucleic acids by over fixation. 15. Since formalin is composed by a proportion of water, which is immiscible with paraffin, a step of dehydration should be performed. 16. Alcohols are either not miscible with paraffin, so they should be cleared from the sample with xylene. 17. There are different protocols applying different paraffin mixtures and melting points temperatures (usually from 55–63  C). Paraffin wax is solid at RT, but once melted, it can infiltrate the tissue. 18. Contrary to the standardized methods, we strongly recommend not to use organic solvents such as xylene for the deparaffination. In addition to the effect on the sample (if the washing steps with ethanol cannot completely remove it), the hazards of xylene for the individual are important if this is inhaled. It is very well documented to be toxic mainly at the level of central nervous system, but also in lungs, liver and kidney, blood, skin, and other systems [9]. 19. The use of Ultra-Turrax is the best option to homogenize the tissue after standard xylene-mediated deparaffination when the sample is immersed in lysis buffer from standard kits. In case of using mineral oil for deparaffination, we do not recommend the use of Ultra-Turrax because of the bubbles problem generated when mixing the lysis buffer and the oil. 20. 80% ethanol for washing steps should be freshly prepared to ensure efficient cleaning of the samples. Before adding DNaseI or RNase-A, respectively, be sure that these treatments will not interfere with downstream analysis as happens for fusion analysis by RNA-seq in case of EwS. References 1. Hewitt SM, Lewis FA, Cao Y et al (2008) Tissue handling and specimen preparation in surgical pathology: issues concerning the recovery of nucleic acids from formalin-fixed, paraffinembedded tissue. Arch Pathol Lab Med 132:1929–1935. https://doi.org/10.1043/ 1543-2165-132.12.1929 2. Suzuki T, Ota C, Fujino N et al (2019) Improving the viability of tissue-resident stem cells

using an organ-preservation solution. FEBS Open Bio 9(12):2093–2104. https://doi.org/ 10.1002/2211-5463.12748 3. Dash A, Maine IP, Varambally S et al (2002) Changes in differential gene expression because of warm ischemia time of radical prostatectomy specimens. Am J Pathol 161:1743–1748. https://doi.org/10.1016/S0002-9440(10) 64451-3

38

Laura Romero-Pe´rez and Thomas G. P. Gru¨newald

4. Vincek V, Nassiri M, Nadji M, Morales AR (2003) A tissue fixative that protects macromolecules (DNA, RNA, and protein) and histomorphology in clinical samples. Lab Invest 83:1427–1435 5. Stanta G, Mucelli SP, Petrera F et al (2006) A novel fixative improves opportunities of nucleic acids and proteomic analysis in human archive’s tissues. Diagn Mol Pathol 15:115–123 6. Fox CH, Johnson FB, Whiting J, Roller PP (1985) Formaldehyde fixation. J Histochem Cytochem 33:845–853. https://doi.org/10. 1177/33.8.3894502 7. Florell SR, Coffin CM, Holden JA et al (2001) Preservation of RNA for functional genomic

studies: a multidisciplinary tumor bank protocol. Mod Pathol 14:116–128. https://doi.org/10. 1038/modpathol.3880267 8. Thavarajah R, Mudimbaimannar VK, Elizabeth J et al (2012) Chemical and physical basics of routine formaldehyde fixation. J Oral Maxillofac Pathol 16:400–405. https://doi.org/10.4103/ 0973-029X.102496 9. Kandyala R, Raghavendra SPC, Rajasekharan ST (2010) Xylene: an overview of its health hazards and preventive measures. J Oral Maxillofac Pathol 14:1–5. https://doi.org/10.4103/ 0973-029X.64299

Chapter 4 Liquid Biopsies in Ewing Sarcoma Manuela Krumbholz and Markus Metzler Abstract Liquid biopsies enable noninvasive therapy monitoring in patients with solid tumors. Specific serum markers such as proteins, hormones, or enzymes released from tumor cells or in response to tumor growth can be used for quantification of the tumor burden. However, only a fraction of pediatric tumors has none of these serum markers, but tumor-specific genetic alterations represent reliable alternatives. Here we describe a method for using genomic fusion sequences as liquid biopsy markers in Ewing sarcoma patients. Key words Liquid biopsy, Serum marker, Genomic fusion sequences, Ewing sarcoma, Cell-free circulating DNA, Droplet digital PCR, Double-quenched probes

1

Introduction Liquid biopsies in Ewing sarcoma enable noninvasive therapy monitoring in addition to imaging techniques, that is, X-ray, magnetic resonance tomography (MRT), or positron emission tomography– computed tomography (PET-CT) [1]. Tumor-specific genomic fusion sequences (e.g., EWSR1-FLI1; present in ~85% of Ewing sarcoma patients or EWSR1-ERG; present ~10% of Ewing sarcoma patients) represent reliable serum markers for quantification of tumor growth [2, 3]. Theoretically, they can be detected in circulating tumor cells (CTCs), exosomal vesicles (tumor derived RNA) or as cell-free circulating tumor DNA (ctDNA) isolated from patient’s blood samples [4]. However, only very low numbers of CTCs in the background of millions of hematopoietic cells are detectable in patients with advanced disease stages [5] and fusion transcript (mRNA) positive exosomal vesicles have only been quantified in cell culture and in Ewing sarcoma mouse models in detectable copy numbers so far [6, 7]. In contrast, genomic EWSR1-FLI1 or EWSR1-ERG ctDNA fusion sequences have recently been presented as reliable serum markers in Ewing sarcoma patients [1, 8]. Quantification of ctDNA copy numbers enables a more frequent therapy monitoring

Florencia Cidre-Aranaz and Thomas G. P. Gru¨newald (eds.), Ewing Sarcoma: Methods and Protocols, Methods in Molecular Biology, vol. 2226, https://doi.org/10.1007/978-1-0716-1020-6_4, © Springer Science+Business Media, LLC, part of Springer Nature 2021

39

40

Manuela Krumbholz and Markus Metzler

and a better evaluation of imaging data which is an important step toward personalized therapy optimization and improved relapse diagnostics. Here we describe a detailed protocol for plasma sample collection, isolation of cell-free circulating DNA from frozen plasma samples and quantification of patient-specific ctDNA copy numbers. Due to the high fragmentation of ctDNA and the large background of non–tumor-derived cell-free DNA copy numbers, primer-probe assays for therapy assessment must be well optimized and applied in a highly sensitive quantification technique like digital PCR. Furthermore, we recommend the use of double quenched probes for better differentiation between negative and positive samples using quantitative PCR.

2

Materials

2.1 Plasma Sample Collection and Storage

1. EDTA tubes. 2. Centrifuge. 3. 80  C freezer 4. 1.5 ml reaction tubes.

2.2 Isolation of Cell-Free Circulating DNA

1. Centrifuge. 2. Kits for isolation of cell-free nucleic acids: (a) NucleoSpin Plasma XS Kit (Macherey-Nagel). (b) QIAamp MinElute ccfDNA Kit (Qiagen). (c) QIAsymphony Circulating DNA Kit (Qiagen) for isolation with a QIAsymphony SP Instrument. 3. PBS buffer (phosphate-buffered Saline).

2.3 Quantification of ctDNA in Ewing sarcoma Patients

1. Primers-probe assay for quantification of the tumor-specific fusion sequence. 2. Primers-probe assay for quantification of a single copy control gene. 3. QX200 droplet generator (Bio-Rad). 4. QX200 droplet reader. 5. Heat sealer. 6. Pierceable foil for heat seal. 7. PCR thermocycler. 8. Electronic multichannel pipette. 9. ddPCR Supermix for probes (no dUTP). 10. Droplet generator oil for probes. 11. DG8 cartridges and gaskets.

Liquid Biopsies in Ewing Sarcoma

41

12. ddPCR droplet reader oil. 13. 96-well plate 14. 27G needles 15. Quantasoft Analysis Software. 16. Nuclease-free water. 17. Tumor DNA (genomic DNA) for positive control. 18. DNA from healthy individuals (genomic DNA and/or cell-free circulating DNA) for negative control.

3

Methods

3.1 Plasma Sample Collection and Storage

1. Collect blood sample in EDTA-tubes and start processing within the next 6 h (see Note 1). 2. Centrifuge EDTA-tubes at 1200  g for 10 min at room temperature. 3. Carefully transfer the plasma into 1.5 ml reaction tubes (see Note 2). 4. Optional: Centrifuge plasma samples at 11,000  g for 3 min at room temperature to remove residual cells (see Note 3). 5. Store plasma samples at 80  C.

3.2 Isolation of Cell-Free Circulating DNA

1. Thaw plasma samples at room temperature. 2. Isolate cell-free DNA from plasma using the appropriate commercially available kit (see Note 4). 3. Store ctDNA samples at 20  C.

3.3 Quantification of ctDNA in Ewing Sarcoma Patients

Patient-specific genomic EWSR1-FLI1 or EWSR1-ERG fusion sequences are used as highly tumor-specific markers for quantification of the ctDNA in plasma samples (see Note 6). For quantification of minimal amounts of ctDNA copy numbers high-sensitive digital PCR techniques have to be applied. 1. For each patient, design an individual primer-probe assay for the quantification of the ctDNA level by locating the probe directly on the fusion site, and primers at the 50 and 30 side of the genomic breakpoint. The ideal fragment length of the resulting amplification product should be 80–100 bp (see Note 7). 2. Use double quenched probes as they allow a better differentiation between positive and negative samples (see Fig. 1). 3. For droplet digital PCR (ddPCR) use ddPCR Supermix for Probes (no dUTP) to perform primer-probe base quantification of amplicons (see Notes 8 and 9).

42

Manuela Krumbholz and Markus Metzler

FAM

BHQ1 single quencher

ZEN

20000

17500

17500

15000

15000

Amplitude

Amplitude

20000

FAM

12500 10000 7500

10000 7500 5000

2500

2500

patient NTC DNA

double quencher

12500

5000 0

IBFQ

0

patient NTC DNA

Fig. 1 Double quenched probes reduce background signal in quantitative PCR assays. Amplitudes of amplicon positive (patient DNA) and negative (nontemplate control, NTC) samples using identical droplet digital PCR assays but differently quenched probes

4. Pipet the reaction mix according to the manufacturer’s instructions (see Note 10). 5. Prepare a positive control reaction mix using genomic tumor DNA isolated from tumor biopsy (see Note 11). 6. Incubate the DNA at 95  C for 10 min. 7. Prepare negative control reaction mixes using human genomic DNA or cell-free plasma DNA of healthy individuals and nontemplate control (nuclease-free water). 8. Denaturate human genomic DNA at 95  C for 10 min. 9. Before the sample cartridge is placed in the droplet generator remove all air bubbles using 27G needles. 10. After droplet generation, carefully transfer the droplets into a 96-well plate using an electronic multichannel pipette at minimum speed. 11. Perform PCR reaction according manufacturer’s instructions. 12. Incubate the 96-well plate overnight at 4  C to increase the numbers of final accepted droplets (see Note 12). 13. Measure samples at the droplet reader. 14. Copies of genomic fusion sequences can be calculated per ml of plasma or per copies of a single copy control gene (e.g., albumin) (see Note 13).

Liquid Biopsies in Ewing Sarcoma

4

43

Notes 1. Prolonged storage time of blood samples at room temperature results in a lysis of white blood cells and a large increase of wildtype DNA fragments in the plasma. The higher background of wild-type DNA will consequently reduce the sensitivity of subsequent assays for quantification of tumor-specific ctDNA fusion fragments. 2. Plasma samples should be stored in small aliquots to avoid repeated thawing–freezing cycles during use. For reliable ctDNA quantification plasma samples can be thawed and refroze only once. 3. This step is recommended if isolation of cell-free circulating DNA will be performed with column based isolation kits. Residual cells or cell debris that could clog the column will be removed. 4. Various commercially kits are available. We recommend the use of NucleoSpin Plasma XS Kit (Macherey-Nagel) for small plasma volumes ( 500 μl) and the QIAamp MinElute ccfDNA Kit (Qiagen) for larger plasma volumes (500 μl–5 ml). For high-throughput sample preparation (500 μl–4 ml) use the QIAsymphony Circulating DNA Kit with the QIAsymphony SP (Qiagen) instrument. Use the kits according manufacturer’s instructions with following modifications: (a) NucleoSpin Plasma XS Kit (Macherey-Nagel) Use the “High sensitivity protocol for the isolation of DNA from plasma”. l

For final elution of DNA pipet 30 μl instead of 20 μl elution buffer to the bottom of the column and incubate column with loaded elution buffer for 5 min at room temperature before centrifugation (see Note 5).

(b) QIAamp MinElute ccfDNA Kit (Qiagen). l

l

Plasma volumes pathToGeneExpressionFile = “/RNAseq_ RAW.tsv” > pathToBigWigFiles = paste0(“/”, c("S1. H3K27ac.bw", "S2.H3K27ac.bw", "S3.H3K27ac.bw")) > outputFolder = "" > histoneMarkName = "H3K27Ac" > sampleName = > CHIPIN_normalize(path_to_bw=pathToBigWigFiles, type_norm=" quantile", raw_read_count=pathToGeneExpressionFile, sample_ name=sampleName, output_dir=outputFolder, organism="hg19", expression_plot=TRUE,

compute_stat=TRUE,

percentage=0.1,

nGroup=20, histone_mark=histoneMarkName)

Several parameters can be modified, for instance beforeRegionStartLength, afterRegionStartLength, regionBodyLength, binSize are parameters of the “computeMatrix” function from the “deeptools” package called by ChIPIN. Modifying it allows the user to give more or less weight to regions upstream and downstream of genes for performing normalization.

278

Gwenneg Kerdivel and Valentina Boeva

ChIPIN creates several output files: l

NAME_constant_genes.bed: contains a list of regions representing “constant genes” used for normalization;

l

NAME_CPMmeanVSsd.pdf: biplot showing the distribution of average expression values and and standard deviation of expression across samples (“constant genes” are highlighted in red);

l

normalized .BIGWIG files;

l

StatsAfter.txt and StatsBefore.txt: text files with statistics reflecting the success of the normalization procedure;

l

After_Normalization.pdf and Before_Normalization.pdf: files that allow for visualizing the success of the normalization by showing density profiles around transcription start sites of “constant genes” before and after normalization.

l

If the expression_plot parameter is set to TRUE, a .pdf file containing the density profile around the TSS of low, medium and highly expressed genes is created for each sample. This plot is useful to verify the efficiency of the antibody used. For example, it is known that highly expressed genes should show higher density of the histone mark H3K27Ac than lowly expressed ones. The normalized .BIGWIG density files can then be used for any subsequent analysis that involves a comparison of signal intensities between samples.

3.6 Subsequent Analysis

At this point, the ChIP-seq data have been processed to call peaks and create density profiles; densities profiles can be normalized across samples if needed. Depending on the questions asked, different types of subsequent analysis can be performed. We provide several examples of downstream analysis in this section.

3.6.1 Motif Analysis

Motif enrichment analysis provides numerous biological insights [59]. Applied to a ChIP-seq dataset generated for a TF of interest, it allows for identifying yet unknown TF binding motifs using de novo motif discovery and detecting TFs that cooperate or compete with the TF of interest by scanning peaks for known motifs. In the context of Ewing sarcoma, motif analysis allowed for discovery of two types of binding motifs of EF1 (classic ETS family motif and GGAA microsatellite repeat [25]). It also provided insights on EF1 binding partners [34]. Motif discovery performed on regions carrying a specific histone mark allows for identifying TFs with an activator or repressive function in a given cell type. For example, it enables discovery of TFs that bind active enhancers and promoters and represent major transcriptional regulators [15].

ChIP-seq Analysis in Ewing Sarcoma

279

Numerous tools have been designed to perform motif analysis. Among the most widely used tools, we would name ChIPmunk [60], MEME and MEME-ChIP [61] for de novo motif discovery, i-cisTarget [62], AME and CentriMO from the MEME toolkit [61] for known motif enrichment, and Homer [63] and RSAT [64] for both analysis types. More detail is covered in this review [59]. 3.6.2 Annotation of Binding Sites

Binding sites of TFs or regions enriched in histone modifications detected with ChIP-seq usually have to be assigned to genes they potentially regulate. The easiest, although not so accurate way to assign regulatory regions to genes is by choosing the closest transcription start site. This can be done manually using home-made scripts or by using published tools such as ChIPseeker [65], annotatePeaks [66] or ChIPpeakAnno [67]. A more sophisticated method, TIP [68], proposes a probabilistic model that quantitatively measures the regulatory relationships between TFs and target genes. In addition, the commonly used GREAT tool tries to predict functions of noncoding regions by analyzing the annotation of nearby genes [69]. GREAT is easy to use via the website version (http://great.stanford.edu/public/html/); the user has to choose a genome assembly and upload both test and background regions. If no background regions are available, it is possible to use the whole genome as background. The association rules can also be modified according to the user’s preferences.

3.6.3 Super Enhancer (SE) Calling

SEs are regions of the genome containing multiple enhancers that are bound by a set of transcription factors that determine cell identity. SEs can be identified using ChIP-seq experiments targeting the H3K27ac histone mark [70], mediator protein MED1 [71] or bromodomain protein BRD4 [72, 73]. Here, we will show how to identify SEs using the LILY tool based on H3K27ac profiles [15]. LILY replicates the ROSE algorithm originally developed by the Young lab [71], but in addition to ROSE, the LILY method includes explicit correction for copy number variation inherent to cancer samples. To run LILY based on the HMCan output, run the following in the command line: $ cat PATHTOLILY/runLILY.R | R --slave --args SAMPLE OUTPUT 12500 2500 REF_refseq.ucsc REF.size

SAMPLE: prefix for the sample of interest containing the path to the HMCan output for the corresponding sample. OUTPUT: path where the results of LILY will be written. 12,500: peak stitching distance (bp). 2500: distance around gene TSSs to annotate gene promoters (bp).

280

Gwenneg Kerdivel and Valentina Boeva

REF.refseq.ucsc: file with transcriptome information. REF.size: file with chromosome lengths used to convert .WIG to . BIGWIG files. The output of LILY is a .BED file “SAMPLE.scores.bed”, where SEs are sorted according to their strengths. Ultimately, SE regions can be used to investigate Core Regulatory Circuitries using COLTRON [74] or CRCmapper [75] tools. 3.6.4 Conversion Between Genome Builds

In certain situations, one may have to compare ChIP-seq results obtained with one genome build (for example, human genome build hg38) to results obtained with a different build (for example hg18 or hg19). Indeed, the read files (.FASTQ or .BAM) are not always available and thus it may not be possible to realign the reads on the genome build of interest and redo the peak calling. In this case, it is possible to convert genomic coordinates from one build to another using the LiftOver tool. This can be done online (http://genome-euro.ucsc.edu/cgi-bin/hgLiftOver) or in the command line as follows: $ liftOver PEAKS.hg18.bed hg18ToHg19.over.chain.gz PEAKS.hg19. bed unlifted.bed

This command converts the coordinates of ChIP-seq peaks obtained on hg18 (.BED file PEAKS.hg18.bed) to coordinates on hg19 (.BED file PEAKS.hg19.bed). The positions that fail to be transformed are stored in the file “unlifted.bed”. It is also possible to convert Wig/BigWig files from one build to another even though it is a rather complicated procedure, which we do not recommend. In this case, the .BIGWIG file should first be converted into the .BEDGRAPH format using the bigwigToBedGraph command, then lifted as previously described to the new genome build and ultimately converted back from .BEDGRAPH to .BIGWIG using bedGraphToBigWig. If the density profile is in .WIG format, it should first be converted to .BIGWIG using wigToBigWig. $ wigToBigWig PEAKS.hg18.wig hg18.size PEAKS.hg18.bw $ bigwigToBedGraph PEAKS.hg18.bw PEAKS.hg18.bedGraph $ liftOver PEAKS.hg18.bedGraph hg18ToHg19.over.chain.gz PEAKS. hg19.bed unlifted.bed $ bedGraphToBigWig PEAKS.hg19.bed hg19.size PEAKS.hg19.bw

ChIP-seq Analysis in Ewing Sarcoma

4

281

Notes 1. Computing requirements: The running time can be decreased by using more powerful configurations or working on a highpowered server, especially when dealing with a large number of samples. One should also be aware of the disk space available on the computer for data analysis as the .BAM files generated at the mapping step can be relatively large; moreover, keeping all intermediate analysis files can significantly increase space requirements. Once the peak calling has been performed, all subsequent analysis steps can be performed at a standard desktop or laptop computer. We recommend the Linux operating system to facilitate the use of command line tools and R. 2. Nonexhaustive list of reads preprocessing tools: fastx_clipper (http://hannonlab.cshl.edu/fastx_toolkit/), Btrim [76] (http://graphics.med.yale.edu/trim/), Trimmomatic [77] (http://www.usadellab.org/cms/?page¼trimmomatic), AlienTrimmer [78] (ftp://ftp.pasteur.fr/pub/gensoft/pro jects/AlienTrimmer/), Cutadapt (https://pypi.org/project/ cutadapt/), and AdapterRemoval [79] (https://github.com/ MikkelSchubert/adapterremoval). 3. ChIP-seq reads: Paired-end reads are preferred to single end reads. Each ChIP-seq sample should be paired with a matched control sample: a whole cell extract sample (WCE, or “input”) or a mock ChIP reaction background sample (IgG control). 4. ENCODE blacklist: This list represents a set of regions in the human and other genomes that exhibit anomalous, unstructured, or high signal in all sequencing experiments independently of cell line or experiment (https://github. com/Boyle-Lab/Blacklist/tree/master/lists). 5. Nonexhaustive list of read mappers: BWA [38] (http://bio-bwa. sourceforge.net/), bowtie2 [80] (http://bowtie-bio. sourceforge.net/bowtie2/), STAR [81] (https://github. com/alexdobin/STAR), and NovoAlign (http://www. novocraft.com/products/novoalign/).

Acknowledgments This work was funded through institutional support from Centre National de la Recherche Scientifique, Institut National de la Sante´ et de la Recherche Me´dicale, ATIP-Avenir, the ARC Foundation (ARC-RAC16002KSA-R15093KS), the SIRIC CARPEM (No. INCA-DGOS-INSERM_12561), and the “Who Am I?” Laboratory of Excellence ANR-11-LABX-0071, funded by the French Government through its Investissement d0 Avenir program, operated by the French National Research Agency (ANR-11-IDEX-0005-02).

282

Gwenneg Kerdivel and Valentina Boeva

References 1. Audia JE, Campbell RM (2016) Histone modifications and cancer. Cold Spring Harb Perspect Biol 8:a019521 2. Cosgrove MS, Wolberger C (2005) How does the histone code work? Biochem Cell Biol 83:468–476 3. Suganuma T, Workman JL (2011) Signals and combinatorial functions of histone modifications. Annu Rev Biochem 80:473–499 4. Josling GA, Selvarajah SA, Petter M et al (2012) The role of bromodomain proteins in regulating gene expression. Genes (Basel) 3:320–343 5. Connelly KE, Weaver TM, Alpsoy A et al (2019) Engagement of DNA and H3K27me3 by the CBX8 chromodomain drives chromatin association. Nucleic Acids Res 47:2289–2305 6. Reiter F, Wienerroither S, Stark A (2017) Combinatorial function of transcription factors and cofactors. Curr Opin Genet Dev 43:73–81 7. Groth A, Rocha W, Verreault A et al (2007) Chromatin challenges during DNA replication and repair. Cell 128:721–733 8. Gilmour DS, Lis JT (1984) Detecting proteinDNA interactions in vivo: distribution of RNA polymerase on specific bacterial genes. Proc Natl Acad Sci U S A 81:4275–4279 9. Gilmour DS, Lis JT (1985) In vivo interactions of RNA polymerase II with genes of Drosophila melanogaster. Mol Cell Biol 5:2009–2018 10. Solomon MJ, Larsen PL, Varshavsky A (1988) Mapping proteinDNA interactions in vivo with formaldehyde: evidence that histone H4 is retained on a highly transcribed gene. Cell 53:937–947 11. Beischlag TV, Prefontaine GG, Hankinson O (2018) ChIP-re-ChIP: co-occupancy analysis by sequential chromatin immunoprecipitation. In: Visa N, Jorda´n-Pla A (eds) Chromatin immunoprecipitation: methods and protocols. Springer, New York, NY, pp 103–112 12. Zhang J, Dominguez-Sola D, Hussein S et al (2015) Disruption of KMT2D perturbs germinal center B cell development and promotes lymphomagenesis. Nat Med 21:1190–1198 13. Alekseyenko AA, Walsh EM, Wang X et al (2015) The oncogenic BRD4-NUT chromatin regulator drives aberrant transcription within large topological domains. Genes Dev 29:1507–1523 14. Rui L, Emre NCT, Kruhlak MJ et al (2010) Cooperative epigenetic modulation by cancer amplicon genes. Cancer Cell 18:590–605 15. Boeva V, Louis-Brennetot C, Peltier A et al (2017) Heterogeneity of neuroblastoma cell

identity defined by transcriptional circuitries. Nat Genet 49:1408–1413 16. Funato K, Major T, Lewis PW et al (2014) Use of human embryonic stem cells to model pediatric gliomas with H3.3K27M histone mutation. Science 346:1529–1533 17. Drelon C, Berthon A, Mathieu M et al (2016) EZH2 is overexpressed in adrenocortical carcinoma and is associated with disease progression. Hum Mol Genet 25(13):2789–2800 18. Kim KH, Roberts CWM (2016) Targeting EZH2 in cancer. Nat Med 22:128–134 19. Kelly AD, Issa J-PJ (2017) The promise of epigenetic therapy: reprogramming the cancer epigenome. Curr Opin Genet Dev 42:68–77 20. Yashiro-Ohtani Y, Wang H, Zang C et al (2014) Long-range enhancer activity determines Myc sensitivity to notch inhibitors in T cell leukemia. Proc Natl Acad Sci U S A 111: E4946–E4953 ˜ ez DG, Kraft K, Heinrich V et al (2015) 21. Lupia´n Disruptions of topological chromatin domains cause pathogenic rewiring of gene-enhancer interactions. Cell 161:1012–1025 22. Braun BS, Frieden R, Lessnick SL et al (1995) Identification of target genes for the Ewing’s sarcoma EWS/FLI fusion protein by representational difference analysis. Mol Cell Biol 15:4623–4630 23. Sand LGL, Szuhai K, Hogendoorn PCW (2015) Sequencing overview of Ewing sarcoma: a journey across genomic, Epigenomic and transcriptomic landscapes. Int J Mol Sci 16:16176–16215 24. Kovar H (2010) Downstream EWS/FLI1— upstream Ewing’s sarcoma. Genome Med 2:8 25. Guillon N, Tirode F, Boeva V et al (2009) The oncogenic EWS-FLI1 protein binds in vivo GGAA microsatellite sequences with potential transcriptional activation function. PLoS One 4:e4932 26. Johnson KM, Taslim C, Saund RS et al (2017) Identification of two types of GGAAmicrosatellites and their roles in EWS/FLI binding and gene regulation in Ewing sarcoma. PLoS One 12:e0186275 27. Dallmayer M, Li J, Ohmura S et al (2019) Targeting the CALCB/RAMP1 axis inhibits growth of Ewing sarcoma. Cell Death Dis 10:1–13 28. Gru¨newald TGP, Bernard V, GilardiHebenstreit P et al (2015) Chimeric EWSR1FLI1 regulates the Ewing sarcoma susceptibility gene EGR2 via a GGAA microsatellite. Nat Genet 47:1073–1078

ChIP-seq Analysis in Ewing Sarcoma 29. Riggi N, Knoechel B, Gillespie SM et al (2014) EWS-FLI1 utilizes divergent chromatin remodeling mechanisms to directly activate or repress enhancer elements in Ewing sarcoma. Cancer Cell 26:668–681 30. Lin L, Huang M, Shi X et al (2019) Superenhancer-associated MEIS1 promotes transcriptional dysregulation in Ewing sarcoma in co-operation with EWS-FLI1. Nucleic Acids Res 47:1255–1267 31. Chaturvedi A, Hoffman LM, Jensen CC et al (2014) Molecular dissection of the mechanism by which EWS/FLI expression compromises actin cytoskeletal integrity and cell adhesion in Ewing sarcoma. Mol Biol Cell 25:2695–2709 32. Katschnig AM, Kauer MO, Schwentner R et al (2017) EWS-FLI1 perturbs MRTFB/YAP-1/ TEAD target gene regulation inhibiting cytoskeletal autoregulatory feedback in Ewing sarcoma. Oncogene 36:5995–6005 33. Johnson KM, Mahler NR, Saund RS et al (2017) Role for the EWS domain of EWS/FLI in binding GGAA-microsatellites required for Ewing sarcoma anchorage independent growth. Proc Natl Acad Sci U S A 114:9870–9875 34. Bilke S, Schwentner R, Yang F et al (2013) Oncogenic ETS fusions deregulate E2F3 target genes in Ewing sarcoma and prostate cancer. Genome Res 23:1797–1809 35. Harlow ML, Chasse MH, Boguslawski EA et al (2019) Trabectedin inhibits EWS-FLI1 and evicts SWI/SNF from chromatin in a schedule-dependent manner. Clin Cancer Res 25:3417–3429 36. R Development Core Team (2010) a language and environment for statistical computing: reference index, R Foundation for Statistical Computing, Vienna 37. Ashoor H, He´rault A, Kamoun A et al (2013) HMCan: a method for detecting chromatin modifications in cancer samples using ChIPseq data. Bioinformatics 29:2979–2986 38. Li H, Durbin R (2009) Fast and accurate short read alignment with burrows–wheeler transform. Bioinformatics 25:1754–1760 39. Li H, Handsaker B, Wysoker A et al (2009) The sequence alignment/map format and SAMtools. Bioinformatics 25:2078–2079 40. Robinson JT, Thorvaldsdo´ttir H, Winckler W et al (2011) Integrative genomics viewer. Nat Biotechnol 29:24–26 41. Ernst J, Kellis M (2012) ChromHMM: automating chromatin-state discovery and characterization. Nat Methods 9:215–216 42. Amemiya HM, Kundaje A, Boyle AP (2019) The ENCODE Blacklist: Identification of

283

Problematic Regions of the Genome. Sci Rep 9:1–5 43. Rodrı´guez-Ubreva J, Ballestar E (2014) Chromatin immunoprecipitation. In: Stockert JC, Espada J, Bla´zquez-Castro A (eds) Functional analysis of DNA and chromatin. Humana Press, Totowa, NJ, pp 309–318 44. Wiehle L, Breiling A (2016) Chromatin immunoprecipitation. In: Lanzuolo C, Bodega B (eds) Polycomb group proteins: methods and protocols. Springer, New York, NY, pp 7–21 45. Wardle FC, Tan H (2015) A ChIP on the shoulder? Chromatin immunoprecipitation and validation strategies for ChIP antibodies. F1000Res 4:235 46. Egelhofer TA, Minoda A, Klugman S et al (2011) An assessment of histone-modification antibody quality. Nat Struct Mol Biol 18:91–93 47. Goens G, Rusu D, Bultot L et al (2009) Characterization and quality control of antibodies used in ChIP assays. Methods Mol Biol 567:27–43 48. Landt SG, Marinov GK, Kundaje A et al (2012) ChIP-seq guidelines and practices of the ENCODE and modENCODE consortia. Genome Res 22:1813–1831 49. Kinsey M, Smith R, Iyer AK et al (2009) EWS/FLI and its downstream target NR0B1 interact directly to modulate transcription and oncogenesis in Ewing’s sarcoma. Cancer Res 69:9047–9055 50. Shimizu R, Tanaka M, Tsutsumi S et al (2018) EWS-FLI1 regulates a transcriptional program in cooperation with Foxq1 in mouse Ewing sarcoma. Cancer Sci 109:2907–2918 51. Zhang Y, Liu T, Meyer CA et al (2008) Modelbased analysis of ChIP-Seq (MACS). Genome Biol 9:R137 52. Chen Y, Negre N, Li Q et al (2012) Systematic evaluation of factors influencing ChIP-seq fidelity. Nat Methods 9:609–614 53. Buisine N, Kerdivel G, Sachs LM (2018) De novo transcriptomic approach to study thyroid hormone receptor action in non-mammalian models. In: Plateroti M, Samarut J (eds) Thyroid hormone nuclear receptor: methods and protocols. Springer, New York, NY, pp 265–285 54. Davis-Turak J, Courtney SM, Hazard ES et al (2017) Genomics pipelines and data integration: challenges and opportunities in the research setting. Expert Rev Mol Diagn 17:225–237 55. Liang K, Keles¸ S (2012) Normalization of ChIP-seq data with control. BMC Bioinformatics 13:199

284

Gwenneg Kerdivel and Valentina Boeva

56. Nair NU, Sahu AD, Bucher P et al (2012) ChIPnorm: a statistical method for normalizing and identifying differential regions in histone modification ChIP-seq libraries. PLoS One 7:e39573 57. Taslim C, Wu J, Yan P et al (2009) Comparative study on ChIP-seq data: normalization and binding pattern characterization. Bioinformatics 25:2334–2340 58. Shao Z, Zhang Y, Yuan G-C et al (2012) MAnorm: a robust model for quantitative comparison of ChIP-Seq data sets. Genome Biol 13: R16 59. Boeva V (2016) Analysis of genomic sequence motifs for deciphering transcription factor binding and transcriptional regulation in eukaryotic cells. Front Genet 7:24 60. Kulakovskiy IV, Boeva VA, Favorov AV et al (2010) Deep and wide digging for binding motifs in ChIP-Seq data. Bioinformatics 26:2622–2623 61. McLeay RC, Bailey TL (2010) Motif enrichment analysis: a unified framework and an evaluation on ChIP data. BMC Bioinformatics 11:165 62. Herrmann C, Van de Sande B, Potier D et al (2012) I-cisTarget: an integrative genomics method for the prediction of regulatory features and cis-regulatory modules. Nucleic Acids Res 40:e114 63. Heinz S, Benner C, Spann N et al (2010) Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol Cell 38:576–589 64. Thomas-Chollier M, Sand O, Turatsinze J-V et al (2008) RSAT: regulatory sequence analysis tools. Nucleic Acids Res 36:W119–W127 65. Yu G, Wang L-G, He Q-Y (2015) ChIPseeker: an R/Bioconductor package for ChIP peak annotation, comparison and visualization. Bioinformatics 31:2382–2383 66. Chen X, Xu H, Yuan P et al (2008) Integration of external signaling pathways with the core transcriptional network in embryonic stem cells. Cell 133:1106–1117 67. Zhu LJ, Gazin C, Lawson ND et al (2010) ChIPpeakAnno: a Bioconductor package to annotate ChIP-seq and ChIP-chip data. BMC Bioinformatics 11:237 68. Cheng C, Min R, Gerstein M (2011) TIP: a probabilistic method for identifying

transcription factor target genes from ChIPseq binding profiles. Bioinformatics 27:3221–3227 69. McLean CY, Bristor D, Hiller M et al (2010) GREAT improves functional interpretation of cis-regulatory regions. Nat Biotechnol 28:495–501 70. Hnisz D, Abraham BJ, Lee TI et al (2013) Super-enhancers in the control of cell identity and disease. Cell 155:934–947 71. Whyte WA, Orlando DA, Hnisz D et al (2013) Master transcription factors and mediator establish super-enhancers at key cell identity genes. Cell 153:307–319 72. Love´n J, Hoke HA, Lin CY et al (2013) Selective inhibition of tumor oncogenes by disruption of super-enhancers. Cell 153:320–334 73. Chapuy B, McKeown MR, Lin CY et al (2013) Discovery and characterization of superenhancer-associated dependencies in diffuse large B cell lymphoma. Cancer Cell 24:777–790 74. Lin CY, Erkek S, Tong Y et al (2016) Active medulloblastoma enhancers reveal subgroupspecific cellular origins. Nature 530:57–62 75. Saint-Andre´ V, Federation AJ, Lin CY et al (2016) Models of human core transcriptional regulatory circuitries. Genome Res 26:385–396 76. Kong Y (2011) Btrim: a fast, lightweight adapter and quality trimming program for next-generation sequencing technologies. Genomics 98:152–153 77. Bolger AM, Lohse M, Usadel B (2014) Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30:2114–2120 78. Criscuolo A, Brisse S (2013) AlienTrimmer: a tool to quickly and accurately trim off multiple short contaminant sequences from highthroughput sequencing reads. Genomics 102:500–506 79. Schubert M, Lindgreen S, Orlando L (2016) AdapterRemoval v2: rapid adapter trimming, identification, and read merging. BMC Res Notes 9:88 80. Langmead B, Salzberg SL (2012) Fast gappedread alignment with Bowtie 2. Nat Methods 9:357–359 81. Dobin A, Davis CA, Schlesinger F et al (2013) STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29:15–21

Chapter 22 Epigenetic Analysis in Ewing Sarcoma Jeremy M. Simon and Nicholas C. Gomez Abstract Ewing sarcoma is a highly malignant tumor characterized by a chromosomal translocation that modifies the activity of an ETS family transcription factor. The most prevalent translocation product, EWSR1-FLI1, exploits a permissive and unique chromatin environment of stem cells, and transforms them into an oncogenic state through alterations to gene expression and gene regulatory programs. Though the transformation ability of, and subsequent reliance on EWSR1-FLI1 had been previously described, the advent of genome-wide sequencing technologies allowed for the specific identification of genomic loci and genes targeted by EWSR1-FLI1. Furthermore, the characterization of the chromatin environment in these, and other, cell types could not have been accomplished without the computational and statistical methods that enable large-scale data analysis. Here, we outline in detail the tools and steps needed to analyze genome-wide transcription factor binding and histone modification data (chromatin immunoprecipitation, ChIP-seq), as well as chromatin accessibility data (assay for transposase-accessible chromatin, ATAC-seq) from Ewing sarcoma cells. Our protocol includes a compilation of data quality control metrics, trimming of adapter sequences, reference genome alignment, identification of enriched sites (“peaks”) and motifs, as well as annotation and visualization, using real-world data. These steps should provide a platform on which molecular biologists can build their own analytical pipelines to aid in data processing, analysis, and interpretation. Key words Next-generation sequencing, ChIP-seq, ATAC-Seq, EWSR1-FLI1, H3K27ac, Epigenetics, Bioinformatics, Ewing sarcoma

1

Introduction Ewing sarcoma is a tumor of the bone and soft tissue that primarily affects children and young adults. It is characterized by chromosomal translocations that involve members of the TET family and the ETS family of transcription factors; 80–85% of tumors exhibit an in-frame fusion of EWSR1 and FLI1 (hereafter EWSR1-FLI1) [1, 2]. Translocations with other ETS genes are detected in most of the remaining tumors, yielding similarly functioning fusion proteins [3, 4]. EWSR1-FLI1 exhibits altered binding properties when compared to its DNA-binding parental protein FLI1 despite identical ETS-family DNA binding domains [5]. Instead of localizing to

Florencia Cidre-Aranaz and Thomas G. P. Gru¨newald (eds.), Ewing Sarcoma: Methods and Protocols, Methods in Molecular Biology, vol. 2226, https://doi.org/10.1007/978-1-0716-1020-6_22, © Springer Science+Business Media, LLC, part of Springer Nature 2021

285

286

Jeremy M. Simon and Nicholas C. Gomez

canonical ETS motifs [6], EWSR1-FLI1 is driven to a certain class of microsatellite repeats characterized by a multimerization of the GGAA core of the ETS sequence motif [7–9]. Which specific microsatellites and other GGAA-containing sequences in the genome EWSR1-FLI1 localizes to is further dependent on the underlying organization of chromatin, both in tumor cells as well as the putative cell of origin, the mesenchymal stem cell [10]. In eukaryotes, DNA is organized into chromatin through tight but dynamic interactions with histones, together forming a structure known as the nucleosome [11]. Histones can be posttranslationally modified, particularly on their flexible tails [12], remodeled by incorporating noncanonical proteins [13, 14], or repositioned or displaced during DNA-templated processes such as transcription [15–17]. This reorganization is often a necessary precursor to the binding of transcription factors, which typically can only bind to their cognate sites in accessible regions of chromatin [18]. What makes Ewing sarcoma etiology both unique and interesting is that stem cells, particularly mesenchymal stem cells, exhibit chromatin accessibility over repetitive elements including GGAA-containing microsatellites [9, 10, 19]. What this suggests is that upon translocation, EWSR1-FLI1 can coopt the already permissive chromatin environment of stem cells, recognize and bind to GGAAcontaining repetitive elements, and then reinforce their role as enhancers, driving oncogenic gene expression [5, 9, 10]. Therefore, chromatin accessibility at critical sites specific to stem cells can facilitate EWSR1-FLI1 targeting [10] and influence downstream gene expression and cancer progression [7, 20–23]. Studying genome-wide chromatin accessibility, transcription factor binding, and localization of histone modifications in depth in multiple cell types relied heavily on the use of high-throughput DNA sequencing technologies [24, 25]. The advent of highthroughput sequencing enabled the development and wide adoption of assays such as chromatin immunoprecipitation (ChIP-seq) [26–28], formaldehyde-assisted isolation of regulatory elements (FAIRE-seq) [29, 30], DNase hypersensitivity (DNase-seq) [31– 33], and assay for transposase-accessible chromatin (ATAC-seq) [34, 35]. Each of these experiments yields a billion or more sequenced nucleotides, the fragments of which then need to be assigned to their source coordinates in the reference genome. Then, in order to tell which sequenced fragments represent true transcription factor binding sites, or accessible chromatin regions, or loci with modified histones, an algorithm must separate true enrichment from background signal [36, 37], while minimizing false discoveries. This is broadly true whether studying chromatin or transcription factors from Ewing sarcoma cells or another biological system. Every one of these steps requires intense computation; however, molecular biologists interested in analyzing their own data can do so by utilizing the example protocol we detail below. We stress

Analysis of Epigenetic Alterations in Ewing Sarcoma

287

that each of the outlined steps are flexible and modular such that other algorithms can be easily substituted if a given user or lab prefers, and that the best methodology will depend on the underlying dataset and hypotheses.

2

Materials/Applications This chapter requires the installation of specific applications and functions to work. Due to the specifics and complexities of different operating systems this chapter will not discuss how to install each package but instead the user is referred to the extensive documentation that each of the respective applications provides. The required applications are listed in Tables 1 and 2.

3

Methods This chapter will utilize the publicly available ATAC-seq and ChIPseq data generated in from the study performed by Riggi et al. [9]. Specifically, we will process and analyze the data generated from SK-N-MCs, a Ewing sarcoma cell line. The first step is to download the data from the NCBI Gene Expression Omnibus (GEO) using SRA Toolkit. The raw reads will be trimmed of sequencing adapters and processed through FASTQC for quality control. Trimmed reads are then aligned to the human genome (hg38) and regions of enrichment are identified by calling peaks. Peaks are then scanned for statistically significant enrichment of transcription factor motifs and associated with genomic features such as promoters, introns, and exons. We then create a heatmap displaying ATAC and H3K27ac signal at EWSR1-FLI1 binding sites to demonstrate a simple option for integrating different types data sets. To perform the following commands please use a commandline UNIX/LINUX interface, Terminal for Mac OS X, or Cygwin for Windows.

3.1 Data Acquisition and Preprocessing

1. Download SK-N-MC Data from GEO using SRA toolkit (see Note 1). #ATAC-Seq fastq-dump --split-files SRX718183 #ChIP-seq H3K27ac fastq-dump --split-files SRX718118 #ChIP-Seq FLI1 fastq-dump --split-files SRX718117

288

Jeremy M. Simon and Nicholas C. Gomez

Table 1 Required Programs/Applications Software

Ver.

Website

ReferenceS

SRA toolkit

2.10.0

https://github.com/ncbi/sra-tools

Skewer

0.2.2

https://github.com/relipmoc/skewer

[38]

FASTQC

0.11.8

https://www.bioinformatics.babraham.ac.uk/projects/fastqc/

[39]

Bowtie2

2.3.5.1

http://bowtie-bio.sourceforge.net/bowtie2/index.shtml

[40]

Samtools

1.9

http://www.htslib.org/

[41]

Samblaster

0.1.24

https://github.com/GregoryFaust/samblaster

[42]

Homer

4.11

http://homer.ucsd.edu/homer/index.html

[43]

R

3.5.1

https://www.r-project.org/

[44]

R studio

1.0.153

https://rstudio.com/

[45]

MACS2

2.2.5

https://github.com/taoliu/MACS

[37]

deepTools

3.3.1

https://deeptools.readthedocs.io/en/develop/

[46]

IGV

2.7.2

http://software.broadinstitute.org/software/igv/

[47]

Table 2 Required Bioconductor Packages—available from Bioconductor.org Package

Ver.

Website

References

Bioconductor

3.8

https://www.bioconductor.org/install/

[48]

ATACseqQC

1.6.4

https://www.bioconductor.org/packages/release/bioc/ html/ATACseqQC.html

[49]

Rsamtools

1.34

https://www.bioconductor.org/packages/release/bioc/ html/Rsamtools.html

[50]

BSgenome. Hsapiens. UCSC.hg38

1.4.1

https://www.bioconductor.org/packages/release/data/ [51] annotation/html/BSgenome.Hsapiens.UCSC.hg38.html

GenomicRanges

1.34.0 https://www.bioconductor.org/packages/release/bioc/ html/GenomicRanges.html

[52]

ChIPpeakAnno

3.16.0 https://www.bioconductor.org/packages/release/bioc/ html/ChIPpeakAnno.html

[53]

Rtracklayer

1.42.1 https://www.bioconductor.org/packages/release/bioc/ html/rtracklayer.html

[54]

Analysis of Epigenetic Alterations in Ewing Sarcoma

289

2. Trim Adapter from sequencing reads using Skewer [Optional but recommended] (see Note 2). 3. Skewer requires a FASTA file of known adapter sequences. We can create this file from the known adapter sequences obtained from the Illumina support site (see Note 3). 4. Run Skewer using 6 threads (-t 6), give the output a new base name (-o SKNMC_ATAC), run in the paired end mode (-m pe), and use the adapter FASTA reference file (-x). skewer -t 6 -o SKNMC_ATAC -m pe -x Skewer_Nextera_adapter_ sequences.fa SRX718183_1.fastq SRX718183_2.fastq

5. Trim adapters from H3K27ac ChIP Seq (see Note 4). skewer -t 6 -o SKNMC_H3K27ac -m tail -x Skewer_TruSeq_ adapters.fa SRX718118_1.fastq

6. Trim adapters from FLI1 ChIP Seq. skewer -t 6 -o SKNMC_FLI1 -m tail -x Skewer_TruSeq_adapters.fa SRX718117_1.fastq

7. Quality Control using FASTQC [Optional] (see Note 5). Each FASTQ file will be processed independently through the FASTQC software (see Note 6). #ATAC-Seq fastqc --noextract --outdir FASTQC_ATAC_1 SKNMC_ATAC-trimmedpair1.fastq fastqc --noextract --outdir FASTQC_ATAC_2 SKNMC_ATAC-trimmedpair2.fastq #H3K27ac ChIP-Seq fastqc --noextract --outdir FASTQC_H3K27ac SKNMC_H3K27actrimmed.fastq #FLI1 ChIP-Seq fastqc --noextract --outdir FASTQC_FLI1 SKNMC_FLI1-trimmed. fastq

3.2 Alignment of Data

1. Align to Human Genome (GRCh38): use bowtie2 to align to the human genome (hg38) (see Note 7) setting the options to very-sensitive, maximum fragment length of 2000 (-X 2000) and number of threads to 8 (-p). #ATAC-Seq bowtie2 -p 8 --very-sensitive -X 2000 --rg-id SKNMC_ATAC -x Bowtie2Index/genome

-1

SKNMC_ATAC-trimmed-pair1.fastq

-2

SKNMC_ATAC-trimmed-pair2.fastq -S SKNMC_ATAC.sam #H3K27ac ChIP-Seq bowtie2 -p 8 --very-sensitive --rg-id SKNMC_H3K27ac -x Bow-

290

Jeremy M. Simon and Nicholas C. Gomez tie2Index/genome

-U

SKNMC_H3K27ac-trimmed.fastq

-S

SKNMC_H3K27ac.sam #FLI1 ChIP-Seq bowtie2 -p 8 --very-sensitive --rg-id SKNMC_FLI1 -x Bowtie2Index/genome -U SKNMC_FLI1-trimmed.fastq -S SKNMC_FLI1.sam

2. Sort and remove duplicates from the datasets (see Note 8). The ATAC-seq data needs to be first sorted by read-pair and then finally stored as compressed version and indexed (see Note 9). #ATAC-Seq #Sort by read pair in order to remove duplicates samtools sort -n -o SKNMC_ATAC.sortedName.sam SKNMC_ATAC.sam #Remove duplicates samblaster -i SKNMC_ATAC.sortedName.sam -o SKNMC_ATAC.sortedName.DupsRemoved.sam -r #Convert to bam and sort by coordinate samtools view -bS SKNMC_ATAC.sortedName.DupsRemoved.sam | samtools sort -o SKNMC_ATAC.DupsRemoved.sorted.bam #Create the index samtools index SKNMC_ATAC.DupsRemoved.sorted.bam #H3K27ac ChIP-Seq #Sort by coordinate samtools

view

-bS

SKNMC_H3K27ac.sam

|

samtools

sort

-o

sort

-o

SKNMC_H3K27ac.sorted.bam #Remove duplicates samtools rmdup -s SKNMC_H3K27ac.sorted.bam SKNMC_H3K27ac.DupsRemoved.sorted.bam #Create the index samtools index SKNMC_H3K27ac.DupsRemoved.sorted.bam #FLI1 ChIP-Seq samtools

view

-bS

SKNMC_FLI1.sam

|

samtools

SKNMC_FLI1.sorted.bam #Remove duplicates samtools rmdup -s SKNMC_FLI1.sorted.bam SKNMC_FLI1.DupsRemoved.sorted.bam #Create the index samtools index SKNMC_FLI1.DupsRemoved.sorted.bam

Analysis of Epigenetic Alterations in Ewing Sarcoma

291

3. Shift ATAC-seq reads to account for transposition event (see Note 10). Rscript Shift_Bam_Files.R SKNMC_ATAC.DupsRemoved.sorted.bam shifted/

3.3

Peak Calling

1. The reads are now aligned to the human genome and duplicate reads have been removed. 2. We can now call peaks on the data sets to identify regions of significant signal enrichment. #ATAC macs2 callpeak -f BAMPE -t shifted/SKNMC_ATAC.DupsRemoved. sorted_shifted.bam -g hs --outdir ATAC_MACS2_Peaks -n SKNMC_ ATAC #H3K27ac macs2 callpeak -t SKNMC_H3K27ac.sorted.bam -g hs --outdir H3K27ac_MACS2_Peaks -n SKNMC_H3K27ac #FLI1 macs2 callpeak -t SKNMC_FLI1.sorted.bam -g hs --outdir FLI1_ MACS2_Peaks -n SKNMC_FLI1

MACS2 outputs: (a) an excel readable file with information about the peak calls, (b) a .narrowPeak which is a BED 6+ 4 which can be loaded into IGV or the UCSC Genome Browser for visualization (below). (c) a .summits file which contains the bp of the summit of each peak and. (d) a model.R file which will create an image of the model for your data. 3.4

Visualization

3.5 CREATE HEATMAPS

There are many ways to visualize the data but common tools include the Integrative Genomics Viewer (IGV) or UCSC Genome Browser [55]. The sorted bam files from above can be loaded directly into IGV in order to visualize the alignments (Fig. 1) (see Note 11). A common analysis is to create heatmaps of various genomic signals at different loci. In this section we will take the EWSR1-FLI1 peaks called from above and plot the ATAC and H3K27ac signal at these regions. This analysis is highly flexible in terms of data sets, genomic regions, and even visualization options. Therefore, the processing steps should be edited to fit the specific needs of the project.

292

Jeremy M. Simon and Nicholas C. Gomez

Fig. 1 IGV Snapshot of Genomic Data. A Browser shot demonstrating ATAC-seq signal (top), FLI1 ChIP-seq (middle), and H3K27ac ChIP-seq data at the Cholecystokinin (CCK) locus. We can see ATAC signal and FLI1 signal at the transcription start site as well as H3K27ac signal flanking the FLI1 binding site

We recommend using deepTools which is a suite of python tools that, among many other useful functions, creates and visualizes heatmaps. 1. deepTools works primarily with bigWig files which are used to display dense continuous data such as alignments along the genome. Therefore, the bam files will need to be first converted into bigWig (bw) files. -b specifies the input file to be converted, -o is the new bw being created, -of is the format requested (bigwig), -bs is the bin size of the file which we set to 1 to give single-bp resolution, and -p is the number of threads to use (see Note 12). #Convert ATAC to bigWig bamCoverage -b shifted/SKNMC_ATAC_DupsRemoved. sorted_shifted.bam

-o

SKNMC_ATAC_DupsRemoved.

sorted_shifted.bw -of bigwig -bs 1 -p 8 #Convert H3K27ac to bigWig bamCoverage

-b

SKNMC_H3K27ac.DupsRemoved.

sorted.bam -o SKNMC_H3K27ac.DupsRemoved.sorted. bw -of bigwig -bs 1 -p 8 #Convert FLI1 to bigWig bamCoverage bam

-o

-b

SKNMC_FLI1.DupsRemoved.sorted.

SKNMC_FLI1.DupsRemoved.sorted.bw

-of

bigwig -bs 1 -p 8

2. In order to compute the matrix for the heatmap we need to give deepTools a bedfile of the regions we are interested in. In this case, we will modify the FLI1 narrowPeak file into the proper format by using the cut command (see Note 13). cut -f 1-6 FLI1_MACS2_Peaks/SKNMC_FLI1_peaks.narrowPeak > SKNMC_FLI1_peaks.bed

Analysis of Epigenetic Alterations in Ewing Sarcoma

293

3. Now we can run the computeMatrix reference-point function from the deepTools suite to create a matrix of scores per genomic region of interest. We indicate the signal file(s) using -S and the regions of interest -R, how far upstream and downstream in basepairs from the reference point to plot (-a, -b), the name of the output (-o), the point of reference in the genomic region (--referencePoint), how to sort the regions in the matrix (--sorRegions, and --sortUsing), and finally the number of threads (-p). computeMatrix

reference-point

DupsRemoved.sorted.bw sorted_shifted.bw

-S

SKNMC_FLI1.

SKNMC_ATAC_DupsRemoved.

SKNMC_H3K27ac.DupsRemoved.

sorted.bw -R SKNMC_FLI1_peaks.bed -a 2000 -b 2000 -o SKNMC_signal_at_FLI1_Peaks.txt.gz --referencePoint center --sortRegions descend --sortUsing mean -p 8.

4. Using the output matrix, SKNMC_signal_at_FLI1_Peaks.txt. gz, we can create a heatmap with the plotHeatmap function. plotHeatmap

-m

SKNMC_signal_at_FLI1_Peaks.txt.gz

-o

SKNMC_signal_at_FLI1_peaks.pdf --refPointLabel Peak_Center --sortUsingSamples 1

5. The resulting pdf plot has our three heatmaps of FLI1, ATAC, and H3K27ac signal at FLI1 peak sites 2 kb from the center of the FLI1 peak (Fig. 2). 3.6 Transcription Factor Motif Enrichment

1. We can look for the presence of statistically enriched transcription factor motifs within our peaks using HOMER (see Note 14). #ATAC Motif Finding findMotifsGenome.pl ATAC_MACS2_Peaks/SKNMC_ATAC_peaks.narrow Peak WholeGenomeFasta/genome.fa ATAC_hg38_HOMER -p 8 #FLI1 Motif Finding findMotifsGenome.pl FLI1_MACS2_Peaks/SKNMC_FLI1_peaks.narrow Peak WholeGenomeFasta/genome.fa FLI1_hg38_HOMER -p 8

2. The output will be an HTML file in which you can read with any browser. The results for FLI1 demonstrate a statistically significant enrichment of the repetitive GGAA motif and predict the EWSR1-FLI1 transcription factor as expected (Fig. 3). 3.7 Genomic Feature Association of Peaks

One common, and useful analysis, is to associate peak calls with specific genomic features. In order to perform these next steps, we will utilize R, R Studio, and various Bioconductor R Packages (see Note 15).

FLI1

ATAC

H3K27ac Signal

10 5

-2.0

Peak_Center 2.0Kb

-2.0

Peak_Center 2.0Kb

-2.0

Peak_Center 2.0Kb

8

7

6

Regions

5

4

3

2

1

-2.0

Peak_Center 2.0Kb

-2.0

Peak_Center 2.0Kb

-2.0

Peak_Center 2.0Kb

0

Fig. 2 Heatmap produced from deepTools. Visualization of FLI1 signal (left), ATAC-seq (middle), and H3K27ac (right) at FLI1 peak sites 2000 bp from the peak center. The data is sorted in descending order by mean FLI1 signal across the window. The scale goes from low (red) to high (blue) signal. We identify coenrichment of FLI1 and ATAC-signal followed by flanking H3K27ac signal consistent with previous reports

Analysis of Epigenetic Alterations in Ewing Sarcoma

295

Fig. 3 Screenshot of HOMER output. HOMER outputs an html file that can be read by any browser. It contains the de novo identified motifs and the putative transcription factor that binds that site. It displays p-values which are recommended to be very significant (